[go: up one dir, main page]

WO2007131213A2 - systèmes et procédés pour une gestion de réputation de supports générés par le consommateur - Google Patents

systèmes et procédés pour une gestion de réputation de supports générés par le consommateur Download PDF

Info

Publication number
WO2007131213A2
WO2007131213A2 PCT/US2007/068308 US2007068308W WO2007131213A2 WO 2007131213 A2 WO2007131213 A2 WO 2007131213A2 US 2007068308 W US2007068308 W US 2007068308W WO 2007131213 A2 WO2007131213 A2 WO 2007131213A2
Authority
WO
WIPO (PCT)
Prior art keywords
response
generated media
consumer generated
user interface
content
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/US2007/068308
Other languages
English (en)
Other versions
WO2007131213A3 (fr
Inventor
Miles Ward
James Webber
Dean Graziano
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Visible Tech Knowledgy Inc
Visible Technologies LLC
Original Assignee
Visible Tech Knowledgy Inc
Visible Technologies LLC
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Visible Tech Knowledgy Inc, Visible Technologies LLC filed Critical Visible Tech Knowledgy Inc
Publication of WO2007131213A2 publication Critical patent/WO2007131213A2/fr
Publication of WO2007131213A3 publication Critical patent/WO2007131213A3/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/951—Indexing; Web crawling techniques
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/958—Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking

Definitions

  • CGM Consumer Generated Media
  • BBS Base Generated Media
  • Product review and discussion web sites online retail sites that support customer comments, social networks, media repositories, audio and video sharing sites/networks and digital libraries.
  • Private non-Internet information systems can host CGM content as well, via environments like Sharepoint, Wiki, Jira, CRM systems, ERP systems, and advertising systems.
  • CCC Consumer created content
  • WSM weblogs and social media
  • WOMM WideMM
  • OWOM Online word of mouth
  • Keyphrase refers to a word, string of words, or groups of words with Boolean modifiers that are used as models for discovering CGM content that might be relevant to a given topic.
  • Boolean modifiers that are used as models for discovering CGM content that might be relevant to a given topic.
  • Post refers to a single piece of CGM content. This might be a literal weblog posting, a comment, a forum reply, a product review, or any other single element of CGM content.
  • Site refers to an Internet site which contains CGM content.
  • Blog refers to an Internet site which contains CGM content.
  • CGM refers to media that resides on CGM sites.
  • CGM is often text, but includes audio files and streams (podcasts, mp3, streamcasts, Internet radio, etc.) video files and streams, animations (flash, Java) and other forms of multimedia.
  • UI refers to a User Interface, that users interact with computer software, perform work, and review results.
  • IM refers to an Instant Messenger, which is a class of software applications that allow direct text based communication between known peers.
  • Thread refers to an "original” post and all of the comments connected to it, present on a blog or forum. A discussion thread holds the information of content display order, so this message came first, followed by this, followed by this.
  • Permalink refers to a URL which persistently points to an individual CGM thread
  • the Internet and other computer networks are communication systems. The sophistication of this communication has improved and the primary modes differentiated over time and technological progress. Each primary mode of online communication varies based on a combination of three basic values: privacy and persistence and control.
  • Email as a communications medium is private (communications are initially exchanged only between named recipients), persistent (saved in inboxes or mail servers) but lacks control (once you send the message, you can't take it back, or edit it, or limit re -use of it).
  • Instant messaging is private, typically not persistent (some newer clients are now allowing users to save history, so this mode is changing) and lacks control.
  • Chat rooms are public (again, some are membership based) typically not persistent, and lack control.
  • Blogs and Social Networks are the predominant communications mediums that permit author control. By reducing the cost, technical sophistication, and experience required to create and administer a web site, blogs and other persistent online communication have given an unprecedented amount of editorial control to millions of online authors. This has created a unique new environment for creative expression, commentary, discourse, and criticism without the historical limits of editorial control, cost, technical expertise, or distribution/exposure.
  • This new medium represents a significant challenge for interested parties to comprehensively understand and interact with.
  • Ql 2007 estimates for the number of active, unique online CGM sites range from 50 to 71 million, with growth rates in the hundreds of thousands of new sites per day.
  • PR, Advertising and Marketing businesses and divisions interact with ( ⁇ 1000 TV channels, ⁇ 1000 radio stations, ⁇ 1000 major news publications, ⁇ 10-20 major pundits on any given subject, etc.) this represents a nearly 10,000-fold increase in the number of potential targets for interaction.
  • Businesses and other motivated communicators have come to depend on software that perform Business Intelligence, Customer Relationship Management, and Enterprise Resource Planning tasks to facilitate accelerated, organized, prioritized, tracked and analyzed interaction with customers and other target groups (voters, consumers, pundits, opinion leaders, analysts, reporters, etc.). These systems have been extended to facilitate IM, E-mail, and telephone interactions. These media have been successfully integrated because of standards (jabber, pop3, smtp, pots, imap) that require that all participant applications conform to a set data format that allows interaction with this data in a predictable way.
  • standards jabber, pop3, smtp, pots, imap
  • Blogs and other CGM generate business value for their owners, both on private sites that use custom or open source software to manage their communications, and for massive public hosts. Because these sites can generate advertising revenue, there is a drive by author/owners to protect the content on these sites, so readers/subscribers/peers have to visit the site, and become exposed to revenue generating advertising, in order to participate in/observe the communication. Because of this financial disincentive, there is no unifying standard for blogs which contains complete data. RSS and Atom feeds allow structured communication of some portion of the communication on sites, but are often very incomplete representations of the data available on a given site.
  • CAPTCHAs Completely Automated Public Turing test to tell Computers and Humans Apart
  • email verification mobile phone text message verification
  • password authentication mobile phone text message verification
  • cookie tracking Uniform Resource Locator (URL) obfuscation
  • timeouts Internet Protocol (IP) address tracking.
  • IP Internet Protocol
  • TruCast is a method for management, by way of gathering, storing, analyzing, tracking, sorting, determining the relevance of, visualizing, and/or responding to all available consumer generated media.
  • consumer generated media include web logs or "blogs", mobile phone blogs or “mo-blogs", forums, electronic discussion messages, Usenet, message boards, BBS emulating services, product review and discussion web sites, online retail sites that support customer comments, social networks, media repositories, and digital libraries.
  • Any web hosted system for the persistent public storage of human commentary is a potential target for this method.
  • the system is comprised of a coordinated software and hardware system designed to perform management, collection, storage, analysis, workflow, visualization, and response tasks upon this media. This system permits a unified interface to manage, target, and accelerate interactions within this space, facilitating public relations, marketing, advertising, consumer outreach, political debate, and other modes of directed discourse.
  • FIGURES 1A-1B shows an example system for consumer generated media reputation management
  • FIGURE 2 shows a method for consumer generated media reputation management
  • FIGURE 3 shows a incoming data from collection being delivered to an ingestion system in one embodiment
  • FIGURE 4 is a depiction of one embodiment of a CGM site discovery system
  • FIGURE 5 provides an overview of ingestion in one embodiment
  • FIGURE 6 shows manual scoring in one embodiment
  • FIGURES 7-9 show the smooth transition between user scoring and automated scoring, in one embodiment
  • FIGURE 10 is a depiction of one embodiment of a CGM response engine
  • FIGURES 11-13 show screen shots of a registration and response feature
  • FIGURE 14 shows an example screenshot of the TruCast Login Authentication screen
  • FIGURE 15 shows an example screenshot of a user interface homepage
  • FIGURE 16 shows an example screenshot of an account manager panel
  • FIGURE 17 shows an example screenshot of a user manager panel
  • FIGURE 18 shows an example screenshot of a topic manager panel
  • FIGURE 19 shows an example screenshot of a topic manager panel with the keyphrase tab activated
  • FIGURE 20 shows an example screenshot of a sorting manager
  • FIGURE 21 shows an example screenshot of the sorting manager with the user tab activated
  • FIGURE 22 shows an example screenshot of a scoring manager
  • FIGURE 23 shows an example screenshot of a scoring manager with a new topic creator screenshot activated
  • FIGURE 24 shows an example screenshot of a response manager
  • FIGURE 25 shows an example screenshot of an administrative queue
  • FIGURE 26 shows an example screenshot of a dashboard launcher
  • FIGURE 27 shows an example screenshot of an impact dashboard
  • FIGURE 28 shows an example screenshot of a sentiment dashboard
  • FIGURE 29 shows an example screen shot of a sentiment history dashboard
  • FIGURE 30 shows an example screenshot of an authority map dashboard
  • FIGURE 31 shows an example screenshot of a data drilldown dashboard
  • FIGURE 32 shows an example screenshot of an ecosystem map dashboard
  • FIGURE 33 shows an example screenshot of an ecosystem map zoom out view
  • FIGURE 34 shows an example screenshot of a sentiment summary
  • FIGURE 35 shows an example screenshot of a set of top lists
  • FIGURE 36 shows an example screenshot of reporting
  • FIGURE 37 shows an example screenshot of an aggregate performance dashboard
  • FIGURE 38 shows a system overview in detail.
  • FIG. IA illustrates an example of a suitable computing system environment 100 on which an embodiment of the invention may be implemented.
  • the computing system environment 100 is only one example of a suitable computing environment and is not intended to suggest any limitation as to the scope of use or functionality of embodiments of the invention. Neither should the computing environment 100 be interpreted as having any dependency or requirement relating to any one or combination of components illustrated in the exemplary operating environment 100.
  • Embodiments of the invention are operational with numerous other general- purpose or special-purpose computing-system environments or configurations.
  • Examples of well-known computing systems, environments, and/or configurations that may be suitable for use with embodiments of the invention include, but are not limited to, personal computers, server computers, hand-held or laptop devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed-computing environments that include any of the above systems or devices, and the like.
  • Embodiments of the invention may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer.
  • program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types.
  • Embodiments of the invention may also be practiced in distributed-computing environments where tasks are performed by remote processing devices that are linked through a communications network.
  • program modules may be located in both local- and remote-computer storage media including memory storage devices.
  • an exemplary system for implementing an embodiment of the invention includes a computing device, such as computing device 100.
  • computing device 100 In its most basic configuration, computing device 100 typically includes at least one processing unit 102 and memory 104.
  • memory 104 may be volatile (such as random-access memory (RAM)), non- volatile (such as read-only memory (ROM), flash memory, etc.) or some combination of the two. This most basic configuration is illustrated in FIG. IA by dashed line 106.
  • device 100 may have additional features/functionality.
  • device 100 may also include additional storage (removable and/or non-removable) including, but not limited to, magnetic or optical disks or tape.
  • additional storage is illustrated in FIG. IA by removable storage 108 and non-removable storage 110.
  • Computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer- readable instructions, data structures, program modules or other data.
  • Memory 104, removable storage 108 and non-removable storage 110 are all examples of computer storage media.
  • Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by device 100. Any such computer storage media may be part of device 100.
  • Device 100 may also contain communications connection(s) 112 that allow the device to communicate with other devices.
  • Communications connection(s) 112 is an example of communication media.
  • Communication media typically embodies computer- readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media.
  • modulated data signal means a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal.
  • communication media includes wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, radio- frequency (RF), infrared and other wireless media.
  • RF radio- frequency
  • computer-readable media as used herein includes both storage media and communication media.
  • Device 100 may also have input device(s) 114 such as keyboard, mouse, pen, voice-input device, touch-input device, etc.
  • input device(s) 114 such as keyboard, mouse, pen, voice-input device, touch-input device, etc.
  • Output device(s) 116 such as a display, speakers, printer, etc. may also be included. All such devices are well-known in the art and need not be discussed at length here.
  • System 200 includes an electronic client device 210, such as a personal computer or workstation, that is linked via a communication medium, such as a network 220 ⁇ e.g., the Internet), to an electronic device or system, such as a server 230.
  • the server 230 may further be coupled, or otherwise have access, to a database 240 and a computer system 260.
  • FIG. IB includes one server 230 coupled to one client device 210 via the network 220, it should be recognized that embodiments of the invention may be implemented using one or more such client devices coupled to one or more such servers.
  • each of the client device 210 and server 230 may include all or fewer than all of the features associated with the device 100 illustrated in and discussed with reference to FIG. IA.
  • Client device 210 includes or is otherwise coupled to a computer screen or display 250. As is well known in the art, client device 210 can be used for various purposes including both network- and local-computing processes.
  • the client device 210 is linked via the network 220 to server 230 so that computer programs, such as, for example, a browser, running on the client device 210 can cooperate in two-way communication with server 230.
  • Server 230 may be coupled to database 240 to retrieve information therefrom and to store information thereto.
  • Database 240 may include a plurality of different tables (not shown) that can be used by server 230 to enable performance of various aspects of embodiments of the invention. Additionally, the server 230 may be coupled to the computer system 260 in a manner allowing the server to delegate certain processing functions to the computer system.
  • the methods and systems are implemented by a coordinated software and hardware computer system.
  • This system is comprised of a set of dedicated networked servers controlled by TruCast.
  • the servers are installed with a combination of commercially available software, custom configurations, and custom software.
  • a web server is one of those modules, which exposes a web based client-side UI to customer web browsers.
  • the UI interacts with the dedicated servers to deliver information to users.
  • TruCast The cumulative logical function of these systems results in a system and method referred to as TruCast.
  • the servers could be placed client side, could be shared or publicly owned, could be located together or separately.
  • the servers could be the aggregation of non-dedicated compute resources from a Peer to Peer (P2P), grid, or other distributed network computing environments.
  • the servers could run different commercial applications, different configurations with the same or similar cumulative logical function.
  • the client to this system could be run directly from the server, could be a client side executable, could reside on a mobile phone or mobile media device, could be a plug-in to other Line of Business applications or management systems.
  • This system could operate in a client-less mode where only Application Programming Interface (API) or extensible Markup Language (XML) or Web-Services or other formatted network connections are made directly to the server system.
  • API Application Programming Interface
  • XML extensible Markup Language
  • the custom server- side engine applications could be written in different languages, using different constructs, foundations, architectural methodologies, storage and processing behaviors while retaining the same or similar cumulative logical function.
  • the UI could be built in different languages, using different constructs, foundations, architectural methodologies, storage and processing behaviors while retaining the same or similar cumulative logical function.
  • FIGURE 2 shows a method for consumer generated media reputation management.
  • the TruCast system can be broken down into elements, the elements are, but are not limited to the following: collection, ingestion, analysis, reporting and response.
  • the Collection system gathers the majority of information about all CGM content online. This is a weighted, prioritized goal because TruCast functions in a weighted, prioritized way.
  • This prioritization system is an optionally advantageous element of the collection system, called the Collection Manager.
  • the Collection Manager receives input from internal and external sources about what sites have information of value, weights that information against a set of pre-described and manipulatable co-factors to allow tuning, and prioritizes the execution of collection against those sites.
  • robot In order to collect data from a blog site, an automated web scripting and parsing system called a robot is built.
  • An individual "robot” is a sophisticated, coordinated script which informs a software engine of how to navigate, parse, and return web information. Every web site is comprised of code in one of several popular languages, which software applications called web browsers "render” or convert to a visually appealing "web site”.
  • a robot similar to a browser, interprets site code to render an output. The desired output is not the "web site” that a browser would create, but an XML document, with columns of information about the content stored on a given site.
  • Robots are accessing the code, and not the rendered page, they have access to markup structures in the code which identify where specific content of interest is stored within the code.
  • Robots use navigation based on Document Object Model (DOM) trees, regular expression pattern matching, conditional parsing, pre-coded transformations, mathematical and logical rules, tags, comments, formatting, and probability statistics to extract the specific content
  • TruCast uses from raw web site code. Functions which perform this parsing are abstracted and codified in the robot engine, which is instructed on specific actions by a specific robot script.
  • Robots can be designed to gather data from only an individual site, or made more general to accommodate variation amongst similar sites. Robots parse the gamut non-structured web site code into XML encoded text that meets a predefined data specification of the design. The system, in one embodiment, collects all posts, all comments, and all desired content from every page that a robot visits.
  • Robots are not limited to these methods for content parsing, hierarchical temporal memory analysis, probability-based positive heuristics, and structural inference technologies can be used to make robots are capable of collecting information from a wider variety of sites.
  • Some sites have full-data RSS or Atom feeds (different than the typically truncated feeds), for which a specific set of robots exist.
  • Robots are not limited to web content collection, but represent a scriptable system for parsing and transforming incoming and outgoing data based on predefined rules.
  • FIGURE 3 depicts one embodiment of a CGM data collection system.
  • the first step of this system is to prioritize possible targets for collection.
  • Inputs to this prioritization include, but are not limited to, sites specifically requested by customers (305) and the number of responses the system is written to a given site (310), the number of accounts that find content from this site relevant (315), the total count of relevant content available on the site (320), the date of the most recent post written on the site (325) and the historical performance of the system at gathering content from this site (330).
  • the priority database maintains an updated list of co-factors which are calculated priorities for each site based on these inputs.
  • the Collection manager (340) determines that it has excess bandwidth/resources to execute more robots, it polls the priority database (335) to determine which robots (345) and then executes them. The collection manager also stores the records of robot activity so that it can add this information to the priority database (335). Robots, once launched by the Collection Manager, interface with their targets (350) to return XML-formatted CGM content to the Ingestion system (355).
  • FIGURE 4 is a depiction of one embodiment of a CGM site discovery system.
  • Site discovery is the process of finding the URLS of new blog sites on the Internet.
  • the coordination is performed by the Discovery Robot Manager (372). This system retains performance information of the three methods, and determines what percentage of available resources (cpu time, bandwidth) to spend running each of the three methods in order to discover the most new URLs possible.
  • the Discovery Robot Manager receives input from the Discovery Targets DB (370) which stores all of the information required to execute each of the three methods, most notably the URL targets for each method.
  • This system is fed information from customer or internal research discovered URLs (362) URLs of known search engines (364) URLs found in the post bodies of CGM content (366) and the URLs of the directory pages for each of the major blog hosts (368).
  • Each method uses this information and a script for web interaction, called a robot, to discover new CGM URLs.
  • the first method is called the "Real Estate” method.
  • the Discovery Robot Manager (372) determines that it is efficient to do so, it will launch a Real Estate robot for a specific search engine (374), and supply it with a list of keywords from all account topics which is held in the Discovery Targets DB(370).
  • This robot will visit the search engine and fill in the search form with each keyword, and gather, by way of regular expression pattern extraction, the URLs of the results from the first 4 pages of results.
  • This information will be delivered in XML format to the de-duplicator (388), which will eliminate known URLs, and then be stored in the Collection Prioritization DB (390) for collection.
  • the second method, Site Search is very similar to the Real Estate method, uses the same robots, but behaves in a different way with different input.
  • the Real Estate robots use keywords from the topics in the accounts.
  • the Site Search method has a pre-determined list of keyphrases designed to be representative of the full gamut of discussion on the web.
  • the Discovery Robot Manager (372) collects this information from the Discovery Targets DB (370) and executes a Site Search robot, which searches the input keyphrases to retrieve the first 20 pages of results. Because of the much larger number of searches, these robots are designed to heavily obfuscate and avoid patterned interaction with Search Engine servers.
  • the URLs discovered by Site Search robots are delivered to the de-duplicator (388), and from there to the Collection Prioritization DB(390).
  • Site Search robots can also alternately be sent input URLs that are blog sites instead of search engines. Within this context they will visit every hyperlink on the site, searching for new links to previously-unknown sites. This be delivered as new URL output similar to the other methods.
  • the third method uses different robots to visit the directory listing pages on major CGM hosting engines. These directory pages' URLs are stored in the Discovery Targets DB(370).
  • the Discovery Robot Manager (372) launches a Host Crawl Robot (376) which visits a CGM Host directory page (382) and visits all of the hyperlinks on that page retrieving all of the URLs that are available. This information is sent to the de-duplicator (388) and on to the Collection Prioritization DB (390) [0078] Ingestion
  • FIGURE 5 depicts one embodiment of a data ingestion system.
  • This system receives input from the XML data outputs of robots launched and administered by the Collection Manager (800). These XML data sources are queued in an Ingestion Queue (805). This queuing process is a buffering function because all of the remaining steps are a stream processing method which requires a steady stream of content to work at maximum efficiency. Due to the dynamic nature of the volume of XML data input, the Ingestion queue holds a backlog of incoming data and outputs it at a steady rate, currently 500 docs/second.
  • This flow of data is delivered first to a system which compares incoming CGM content information to all previously collected content, based on posted date, permalink URL, and post body to ensure that the data does not already exist in the system.
  • This is the de- duplicator (810).
  • this system has culled duplicate documents, it hands those documents to a UREF constructor (815) which creates a new uniquelD number to easily index and track unique content within the system in one embodiment of the invention.
  • content is delivered to a GMT time aligner, which converts all date and time stamps to be relative to Greenwich Mean Time (820).
  • this XML format information is transformed using an XSLT (825) or extensible Style Lanugage Transformation processor, which reformats the data for rapid delivery into the indexing system and relational DB systems (830).
  • TruCast performs several cleaning and refining steps upon incoming CGM content enclosed in the XML format.
  • the system eliminates duplicate content using a fuzzy logic comparison between existing stored content and incoming new content based on post body, permalink, and date information. This comparison is tunable and weighted, where positive matches are clear indicators of duplication, but agreement is required across multiple values to confirm duplication. For example, if two posts came from exactly the same date and time to the second, it's unlikely, but possible, that they are truly different unique posts.
  • this comparison includes text clustering analysis, to use word counts as a computationally inexpensive way to further evaluate uniqueness. Content that is malformed or incomplete according to the data spec is removed and warnings sent to the responsible collection manager element. Once a document is determined to be unique a UREF (unique reference) value is created and appended to it so that there is a relevant single value to index this information within the system. All incoming post dates are aligned to GMT.
  • TruCast delivers all prepared content into an indexing system which formats the data in such a way that it can be rapidly searched based on relationships to other data, keyword presence, account relevance, and date.
  • This structure includes storage of data within a distributed indexed data repository as well as several SQL databases. Each SQL database is optimized for a different consuming system: the UI, the visualization systems, the reporting and statistics systems, the collection priority database, and the target discovery database, as well as the individual account level data stores.
  • TruCast is designed to determine, with a high degree of confidence, the conceptual relevance of a given piece of CGM content to a "topic" or concept space.
  • Topics can be of any breadth ("War” is just as sufficient a topic as "2002 Chevy Silverado Extended Cab Door Hinge Bolt Rust").
  • Topics are abstract identifiers of relevance information about a given piece of CGM content.
  • Each topic can also be understood as a list of "keyphrases" or keywords with Boolean modifiers.
  • Each topic can contain an unlimited number of keyphrases that work as the first tier of pattern matching to identify content that is relevant to an individual account.
  • Each post discovered by the system and, in one embodiment, could be relevant to one topic, many topics, many topics across many accounts, or no topics at all.
  • FIGURE 6 depicts one embodiment of a system for manually appending topic relevance and topical sentiment to blog posts. This process begins by discovery of potentially relevant content by way of keyphrases. Keyphrases are grouped into topics. Topics and keyphrases are created by users (405) in the Topic Manager panel (410) within the UI. Once a new topic and keyphrase is created, this information is transmitted to the indexing system (415) which begins to examine all incoming data for matches against this keyphrase. The information is also handed to the relational database system (420) which is also the StoreDB component of the Historical Data Processor as illustrated in FIGURE 38. This system examines all data that has already been processed to see if it matches this new keyphrase.
  • TruCast contains a user interface that allows users to create topics, create keyphrases that are used to search for potentially relevant posts for that topic, place potentially relevant content into a queue for review, review the text and context of individual content, mark that content as relevant to none, one, or many topics, (thereby capturing human judgment of relevance), and store that information in the relational database.
  • This system is called the Scoring Manager.
  • This method where a post is matched by keyphrase, scored by humans, and delivered to the outputs of TruCast, in one embodiment (visualizations, reports, and response), is the most basic "manual" behavior of the system.
  • FIGURES 7-9 show the smooth transition between user scoring and automated scoring and depict the progression of the operation of one embodiment for an automated categorization and sentiment analysis system. This progression occurs from the early state, where the automated system performs poorly due to a lack of contextual examples, to a mature state where the automated system performs excellently as a result of robust contextual examples.
  • the system in one embodiment, reacts to this improvement by reducing the rate of post queue delivery to users and increasing the acceptance of analyzed posts from the automated system as confidence ratings and exemplar set sizes increase. This process accepts input from the ingestion system (350) into two separate queues.
  • the first queue delivers content to the scoring manager (610) where it is scored by humans (615) and then delivered to the per-topic exemplar sets (620) based on topic relevance, the relational database (625) for storage and use in the response, visualization and report sections, and to an agreement analysis system (645).
  • a second queue delivers content to the automated categorization system which accepts input from the per-topic exemplar sets, as well as topic performance and tuning information from the agreement analysis system (645). This system passes conceptually relevant content to the sentiment analysis systems which also has access to the exemplar and agreement analysis tuning data.
  • the automated systems append a "confidence" score to their evaluations, which are used as a threshold to decide trust in the evaluation's accuracy.
  • TruCast contains a Conceptual Categorization system which has functionality to evaluate posts for relevance by way of statistical analysis on examples provided by humans using the scoring system. Because humans are reviewing the content, from a specific customer's perspective, that content is reliably scored in context. If the above example post (EX.l) was scored by a human scorer for Microsoft, it would be found irrelevant to the Windows Vista operating system. By statistical analysis of hundreds of posts marked relevant or irrelevant to individual topics, the system can utilize not just keywords, but the entire body of the post to determine relevance.
  • TruCast in one embodiment can leverage Optical Character Recognition (OCR) image to text conversion, Fast-Fourier Transform (FFT) and Granular Synthesis (GS) analysis based speech-to-text conversion, as well as Hierarchical Temporal Memory (HTM) processing.
  • OCR Optical Character Recognition
  • FFT Fast-Fourier Transform
  • GS Granular Synthesis
  • HTM Hierarchical Temporal Memory
  • users score content for relevance in the scoring manager they also may assert the sentiment of the content for each topic that it is relevant, from the perspective of their account. Users will mark, from their perspective (as informed by a set of scoring rules described by user administrators) the sentiment reflected about each topic. This information will be stored for later use in a relational database.
  • TruCast collects in one embodiment. This is the "manual" path for data to flow thru the system, in one embodiment,.
  • the second use for these posts is that they serve as example data for an exemplar driven automated sentiment analysis system that mirrors the conceptual categorization system.
  • the system leverages an exemplar set of documents to perform an automated algorithmic comparison in order to determine the sentiment, per topic, contained within an individual post.
  • This requires a larger number of examples than categorization analysis, (-100 per sentiment value per topic) due to the four different stored sentiment values, "good", “bad” , “neutral” and "good/bad”.
  • a lexicon of sentimental terms is stored within the system, and their presence has a weighted impact on the analysis. Negation terms and phrase structures also alter the values associated with sentimental terms.
  • a stop words list eliminates connective terms, object nouns, and other non-sentimental terms within the text, reducing the noise the comparison has to filter thru.
  • Sentence detection uses linguistic analysis to subdivide posts into smaller sections for individual analysis. A series of algorithms are compared for accuracy and performance on a per topic basis, to allow the performance of the analysis system to be tuned to each topic.
  • Both of these processes work upon the post-ingestion content, directing automatically analyzed documents into the remainder of the system workflow. This process reacts to the number of exemplar documents that are available. If incoming content is keyphrase-relevant to a specific topic, a determination is made if sufficient exemplar documents have been gathered by the system from users. If enough exemplary documents are not available, that post is delivered to the scoring queue which feeds content to the scoring manager interface. If some documents are present as exemplars, the system will attempt automated categorization and sentiment analysis, but still deliver posts to the scoring manager. This creates a pair of analysis results, one from the computer and one from the user. These are compared, and when a sufficient alignment (agreement frequency) is reached, the system starts delivering auto-analyzed content directly to the reporting and response systems, saving human effort.
  • the system utilizes an aging and auditing system to ensure that the oldest human scored posts are ejected from the exemplar set and replaced by new human scored posts over time.
  • the system also performs internal cluster analysis and ejects significant outliers from the system. Both of these processes are tunable by administrative control panels.
  • the result of this aging and auditing should be that as the vernacular, word usage, and issues discussed internal to a given topic change over time, exemplar documents continue to reflect that change and accurately map relevance.
  • the system in one embodiment, of databases which receive topic relevant, analyzed content is connected to a series of web- based visualizations to allow users of the UI to understand valuable information about the discussions captured by the system, in one embodiment,. Visualizations are shown in FIGURES 27-38.
  • FIGURE 10 is a depiction of one embodiment of a CGM response engine.
  • the Response Manager UI (752) is populated with a written response by a user (758). This user is evaluated for authorization permissions against a stored value in the Account Database (754). If the user does not have appropriate authorization, their response will be delivered to an authorization queue(756) to be approved by an administrator. If a response is not approved it is deleted. If a responder has authorization, or their response is approved, it will be delivered to the Response Priority Processor (760) which determines if any delay or promotion is required for a given approved post. It also observes the original posted date of the content that is being responded to and prioritizes based on most recent posted dates.
  • the Response Engine Manager (764) requests responses from the Response Priority Processor(760) to deliver to the registration and response robots.
  • the Response engine manager checks the response performance DB (766) to see if a given URL has a response robot that has already been created or not. If it has not, the response and all associated information is sent to the Response Robot Constructor (772).
  • This tool provides an interactive UI to allow semi-automated interaction with a target CGM site's registration and response systems to deliver the response to the site, and record the interaction. These interactions include loading pages, following hyperlinks, assigning input data to site form fields, navigating to web mail systems for authentication messages, completing CAPTCHA tests, interacting with IM and SMS systems, performing sequential interactions in correct order and submitting forms.
  • the result of these actions should be a newly registered user (if required by the site) and a response written to the blog site.
  • the interaction is recorded and stored in the Registration and Response Robot sets(770, 774). If, when the Response Engine Manager is sent a response, it determines that a robot already exists, it will execute that robot without human interaction. This has the same effect, creating a new registration if required, and writing the response to the CGM site. Success or failure of robots and robot constructor actions are recorded in the Response Performance DB for evaluation and manual code rework if required.
  • the response manager is a system to convert into a manageable, scalable business process the task of responding to CGM content by way of comments.
  • All CGM systems that allow interactivity have a web based system for allowing readers of content to respond by way of a comment, note, or other stored message. This often requires that users register themselves on the site, by providing a username, password, and other personal details. Sometimes this requires providing an e-mail address, to which an activation link is sent, or an instant messenger account which is sent a password. This isn't too difficult for casual users to maintain, especially if they only interact with a few sites. Professional users however often have to interact with thousands of different sites.
  • the system in one embodiment, aims to reduce this workload for responders by automating the registration and response process.
  • the TruCast UI system facilitates a workflow for many users to interact in a coordinated, managed way with CGM content. Once a post as been successfully analyzed by either a user in the scoring manager, or the automated analysis systems it becomes available within the response manager. This is a UI system for a user to write a comment in response to relevant posts.
  • the system provides an interface called the response vault for managers to pre-write message components, fragments of text, names, stats, and pieces of argument that they'd like responders to focus on. These snippets can be copied into the response body during authoring. Once a user is done writing a response, the can click a "send" button which delivers the newly written response to the relational database.
  • FIGURES 11-13 show screen shots of a registration and response feature.
  • the system receives a response record from the response manager, it determines which blog site contains the original message, and the link to the response page for that site and message. If the system, in one embodiment, has never written a response to that site before, the record is delivered to the response interactor UI or Response Robot Constructor, which is run by company employees. This UI allows an employee to visit the appropriate site, navigate to the appropriate fields, and assign the information from the record to fields on the site that will cause the site to record a response. This action is recorded, and converted into a script, which is stored as a new robot for later re-use. If TruCast has already written a response to a given site, this script will be used eliminating the need for repeated human interaction.
  • Registration and Response robots are scripted automations, which interpret the code of CGM content pages, web pages, pop3 or web based e-mail systems, and other data structures, and perform pre-determined, probabilistic, or rule driven interactions with those structures. By interpreting page code and scripted instructions, they can imitate the actions of human users of these structures, by executing on screen navigation functions, inserting data, gathering data, and reporting success or failure.
  • An example registration robot would be given as a data input the registration information for an individual user of the system, in one embodiment,, and given the URL to a site that the user wishes to register on.
  • the robot would visit the site, navigate by markers pre-identif ⁇ ed in the page code to the appropriate form locations to insert this information, confirm it's insertion, and report success, as well as any output information from the site.
  • An example response robot would accept as input the registration information for a given user of the system, in one embodiment,, the blog response they've written, and the URL to the site that the user wishes to respond to.
  • the robot would load the site into memory, navigate the page by way of hyperlinks or pre-determined, probabilistic or rule driven information, examine the page source code to discover the appropriate form fields to insert this input data into, do so, and report success.
  • Other embodiments of this solution could include purpose built scripts that perform the same assignment and scripted interaction with CGM sites to perform registration and response tasks. Smaller scale systems would have users perform the manual field entry and navigation tasks, but captures these interactions for conversation involvement identification and maintenance by the analysis systems.
  • the system receives a response record from the response manager, it determines which blog site contains the original message, and the link to the response page for that site and message. If the system, in one embodiment, has never written a response to that site before, the record is delivered to the response interactor UI, which is run by company employees. This UI allows an employee to visit the appropriate site, navigate to the appropriate fields, and assign the information from the record to fields on the site that will cause the site to record a response. This action is recorded, and converted into a script for later re-use. If TruCast has already written a response to a given site, this script will be used eliminating the need for repeated human interaction.
  • This system utilizes a similar engine and scripting methodology as the collection system.
  • Other embodiments of this solution could include purpose built scripts that perform the same assignment and scripted interaction with CGM sites to perform registration and response tasks. Smaller scale systems would have users perform the manual field entry and navigation tasks, but captures these interactions for conversation involvement identification and maintenance by the analysis systems.
  • TruCast is engine and script driven, and each transaction happens by way of a modular execution system, we can tie the process to outside support modules to defeat these automation prevention systems.
  • the response automation system has a complete pop3 e-mail interaction system which can generate e-mail addresses for use in registration, check those addresses for incoming mail, and navigate the mail content as easily as more typical web content.
  • the response automation system uses advanced OCR processing along with human tuning to defeat CAPTCHA protections.
  • the system has access to jabber protocol interactions to create automated IM accounts and interact by SMS with mobile phone systems. TruCast also stores a significant body of information, in contact card format, about responders so more complex registration questions can be correctly answered.
  • the response system within TruCast delivers posts to blog sites, which are the target for the collection system.
  • collects content it matches incoming content to evaluate if that content belongs to a thread that the system has interacted with.
  • the system discovers posts that were written after a response that TruCast wrote, it is returned to the queue of posts assigned to the user who wrote the response, with a maximum priority. This way a conversation can be facilitated.
  • Indicators of origination include: (1) E-mail address used in registration/response process; (2) Owner of e-mail address domain's as reported by the WHOIS information; (3) Receipt of e-mail sent to this address by the correct customer to the system, in one embodiment,; (4) IP Address used in the response/registration process; (5) Reverse DNS lookup on the IP Address used in the response/registration process, and the resultant WHOIS information; and/or (6) Internal consistency of blog user registration information.
  • Any given customer or user will direct a domain name that's appropriate for blog post response, connect this domain (and its MX record) to web accessible server.
  • This server should make available the e-mail addresses hosted on it via a pop3 connection. This resolves issues 1 and 2 by placing ownership of the domain from which the e-mails for registration are generated into the hands of the users.
  • the Response Automation tool receives port 80 from the IP address used for the e-mail server installation, and the server hosts the Response Automation Engine for use in executing the scripting that is created to perform automated response. This resolves issues 4 and 5 by aligning the IP source of the comments with the e-mail source of the comments.
  • the tool collects significantly more information about responders than is typically necessary. This includes obscure information like birth date, favorite car, mother's maiden name, favorite popsicle flavor, user picture, etc, to ensure that registrations are complete, feature rich, and transparent.
  • the manual response app and robots accept this data in the response and registration steps. This resolves issue 6.
  • IP and e- mail anonymization features can be enabled. This obfuscates the source of output messages by way of a rotating IP proxy environment which leverages P2P and onion topologies for maximum opacity.
  • the system has a set of authorization features. Administrators have access to a per-user toggle which forces the posts that users write to be delivered to a review queue instead of the response automation system when they press the "send" button. This queue is accessible by administrators to allow review, editing, or rejection before messages are submitted.
  • Administrators can also create and manipulate sorting rules which prioritize content within user scoring and response queues based on topic, site, engine, author, and date information. This forces users to work on appropriate content, and allows administrators to segment scoring and responding tasks to SME's who have the most context for a given topic, site, engine or author.
  • Users in the system are members of accounts, and afforded permissions within the system based on the role assigned to them by administrative users on a per account basis. Roles are pre-bound permission sets. Administrators can create, edit, and delete everything within the system, except accounts. Group administrators, who have access to multiple accounts, can create accounts, and can edit and delete accounts that they've created or been given access to. System administrators can add, edit, and delete all accounts, so this permission role is reserved for internal support use only. Users within the system, in one embodiment, are intended to perform the majority of the scoring and responding work, and as such have only access to the scoring manager, response manager, and their own user manager to review their own performance. Group users can do these tasks for multiple assigned accounts.
  • Viewers within the system have read only access to all UI controls. Group Viewers can review multiple accounts. Accounts as a whole can be enabled or disabled, which blocks users from accessing the system if their account is disabled, and stops any account specific collection, analysis or processing tasks.
  • FIGURE 15 shows an example screenshot of the user interface homepage 1300.
  • the homepage 1300 enables a user to navigate through the different functions of the UI.
  • the toolbar is located at the bottom of the screen and features two menus (account menu and control panel) and a row of eight icons: Account Manager 1305, User Manager 1310, Topic Manager 1315, Sorting 1320, Scoring Manager 1325, Response Manager 1330, Dashboards 1335, and Reporting 1340.
  • the account manager 1305 is used to create/set-up accounts and deactivate/reactivate accounts.
  • the user manager 1310 is used to set-up/create users, establish group rights and permissions, and to review user activity.
  • the topic manager 1315 is used to set-up/create topics and to set-up/create key phrases.
  • Sorting, 1320 is used to set-up/create scoring and responding rules for a topic, site, author, engine, and/or date and assign rules to a specific user.
  • the scoring manager 1325 is used to read/score posts and create new topics while scoring a post.
  • the response manager 1330 respond to posts in near real time and create/save personas and pre-determined responses.
  • Dashboards 1335 is used to map and graph sentiment, impact, authority and data.
  • Reporting, 1340 is used to display statistical charts.
  • Finally a control panel 1345 is used to log out of TruCast and allows email to be sent directly to user support.
  • FIGURE 16 shows an example screenshot of the account manager 1305.
  • the account manager is accessed by a user through button 1305 in FIGURE 15.
  • the account manager 1305 creates and manages accounts in TruCast. Accounts serve as the logical groups of related users, topics, and other system elements. This creation action establishes a new GUID identified accountID that is used by the backend systems to identify data pertinent to this account. Account is often synonymous with customer for TruCast.
  • FIGURE 17 shows an example screenshot of the user manager 1310.
  • the user manager 1310 allows administrators to set-up users, to assign specific rights/permissions to them and to evaluate their activity in TruCast. This is how a work team is created to address a specific target issue within the CGM space.
  • Each new user is assigned a userID value to track their activities, and identify their actions at the database level, enforce permissions and limit access. All users who login to TruCast already have a userID.
  • the response authorization required flag determines if a user's responses need to be approved by an administrator via the authorization system.
  • FIGURE 18 shows an example screenshot of a topic manager 1315.
  • FIGURE 19 shows an example screenshot of the topic manager 1315 with the Keyphrases tab activated.
  • the Topic Manager 1315 is where administrators define topic titles, create topic descriptions, determine key phrases, and exclude specific phrases from the assigned topic. This will determine the content that is matched by the keyphrase tier of relevance analysis in TruCast. Topics are also analysis points, so they're used later to compare and contrast in the visualization systems. Each topic and keyphrase has a GUID value distinguishing it within the database systems.
  • FIGURE 20 shows an example screenshot of a sorting manager 1320.
  • FIGURE 21 shows an example screenshot of the sorting manager with the users tab activated.
  • Sorting 1320 enables administrators to define scoring and responding guidelines. Administrators can create rules that either impacts all users or a specific user's scoring or responding queue will be sorted. These sorts impact the queue by matching, so all posts that match the rule are sorted to the top of the queue, which allows users to score items that are of general importance after completing scoring the posts that were specifically assigned to them by an administrator.
  • FIGURE 22 shows an example screenshot of a scoring manager 1325.
  • the analysis system having determined that a post matching either keyphrase or conceptual categorization, filtered by the sorting system, delivers posts in a sequential queue to the scoring system. Scoring is the central method for users to impact the function of the automated systems, providing examples and context for their operation and it's the shortest path for a post to make it from ingestion to visualizations and response.
  • the post is placed in text box 2005, the topics that the post relates to are in box 2010, which a user will rate using the radio buttons presented. Finally the site information related to the post is placed in box 2015.
  • FIGURE 23 shows an example screenshot of creating a new topic 2110 in the scoring manager 1325. Because pre-determined topics may not cover the scope or issues that exist in the discussion discovered by TruCast, TruCast allows scoring users to create topics, in the new topic text box 2110, on the fly to capture the observation that a new loci of discussion exists. These topics are not populated with keyphrases at this step. Instead, administrators have the capability to merge and delete topics from the Topic manager to ensure that all the team members who may have simultaneously discovered this new topic can receive direction from the administrator as to what the final topic title will be, and instructions by way of descriptions and scoring rules about how to interpret it.
  • FIGURE 24 shows an example screenshot of a response manager 1330.
  • the output from the analysis system and the scoring manager 1325 feed into the response manager 1330 based on applicable sorting rules as assigned by administrators.
  • Writing the response, in block 2210, and clicking "post" is all that's required to ensure that the message you typed makes it out as a comment on the target site. .
  • Your writing process is supported by significant contextual information, from the topic relevance and sentiment score information to stats about the original author and the site they posted on. Once you submit one response, the next item for your review is available immediately in the same panel, no need to navigate to other pages or sites to find the next place to communicate.
  • FIGURE 25 shows an example screenshot of an administrative queue 2300.
  • the administrative queue tools allow administrators to exercise control over User response activities. These queues can be used for managerial oversight, legal review, tactical analysis, training, feedback and performance auditing. They create the framework for administrative authority over the response process.
  • FIGURE 26 is an example screenshot of a dashboard manager 1335.
  • the Dashboard displays data in dynamic graphical charts and graphs. It maps and reports information based on impact, sentiment, authority and data. This allows users to easily identify critical issues, compare topics of discussion for volume, breadth, depth, tone and interconnectedness of CGM discussions, as well as other useful insights about the CGM space.
  • FIGURE 27 is an example screenshot of an Impact Dashboard 2500 and it refers to a set of three line graphs which show daily totals over time that depicts the breadth, depth, and participation of the discussions contained within one or many topics. This information is combined with a polar chart that shows the combined values of the three graphs for one period.
  • FIGURE 28 is an example screenshot of a Sentiment Dashboard: refers to a snapshot view of a single period, showing the relative post volume versus the average sentiment of your selected topics.
  • FIGURE 29 is an example screenshot of a Sentiment History Dashboard. This display is connected to a history view which displays this information over time.
  • FIGURE 30 is an example screenshot of an Authority Map Dashboard: refers to a node and edge style interactive display which shows the interconnectedness and relative authority of individual authors within a given topic. It shows topic as the center node, sites that contain relevant content as first edge nodes, and authors as second edge nodes. Edges between authors connote comments, links, quotes, and trackbacks as methods of identifying connection and communication. A list view on the right side of the screen allows you to quickly find specific authors or sites within the display. Adjustable level of depth controls allow users to establish constraints (show only authors with more than 2 links, show only positive authors, etc.) that effect the visibility of nodes in the display.
  • FIGURE 31 is an example screenshot of a Data Dashboard: refers to a display that shows a tabular result set of posts that matched the topics selected. This table shows one post per row with columns for date, author name, permalink, site name, sentiment, and topic. This view can show only information based on keyphrase-relevance, or full analyzed relevance, or show those two together. In several other dashboards there are links to more information about a given topic or author. Those links point to this display.
  • FIGURE 32 is an example screenshot of an Ecosystem Map: refers to an Ecosystem map is a node and edge style display of all of the sites that make up the discussion ecosystem for a given topic or topics. It shows a node for each site that contains posts or comments relevant to the topics selected and date ranges selected in the dashboard launcher panel. Between nodes, there should be an edge for each link that connects nodes together.
  • FIGURES 31-32 show example screenshots of Nodes and are size scaled depending on how many posts/relevant posts they have, and colored by average sentiment. Edges are thicker depending on how many links exist between two nodes, and have size scaled arrows showing the predominant direction or ratio of links.
  • Nodes, if clicked on should show the site name, # of posts total, # of relevant posts, and sentiment %. The name is a hyperlink to the site. By selecting an individual topic a more detailed display with the sites and authors most important to a given topic displayed. Double click on the node would lead to the data dashboard with a list of all the titles and permalinks to the relevant posts on that site. Edges, if clicked on, show the # of links represented, % directionality.
  • FIGURE 35 is an example screenshot of a Sentiment Summary: refers to a single topic display that shows the number of authors per sentimental category on a given topic or sum of topics.
  • FIGURE 36 is an example screenshot of Top Lists: This provides users with a set of ranked lists of sites, authors, and posts that are the most relevant, most popular, most negative, most positive, most authoritative, most influential, most linked to, most commented on, or most responded to depending on user selection.
  • FIGURE 37 is an example screenshot of a Reporting:
  • the reporting system provides a series of charts based on selection criteria revolving around CGM content. Daily or total values of posts by keyphrase match or post-analysis match, per topic or topics, site, author, by date range. Performance metrics on scorers and responders are also available, per site, topic, or date range.
  • FIGURE 38 is an example screenshot of an Aggregate Performance Dashboard: This dashboard supplies a cluster of configurable widgets for tracking the relationships between several KPFs associated with the data available within TruCast, in one embodiment.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Finance (AREA)
  • General Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Data Mining & Analysis (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • Information Transfer Between Computers (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

L'invention concerne TruCast qui est un procédé pour la gestion, au moyen de regroupement, stockage, analyse, suivi, tri, détermination de pertinence, visualisation et réponse de tous les supports disponibles générés par un client. Certains exemples de supports générés par un client comprennent des journaux Web ou 'blogues', des blogues de téléphone mobile ou 'blogues-mo', des forums, des messages de discussion électroniques, Usenet, des cartes de message, des services d'émulation BBS, des sites Web de discussion et d'analyse de produits, des sites de vente au détail en ligne qui prennent en charge les commentaires des clients, des réseaux sociaux, des référentiels multimédia et des bibliothèques numériques. Un système quelconque hébergé par le Web pour le stockage public persistant de commentaires humains est une cible potentielle pour ce procédé. Le système est composé d'un système logiciel et matériel coordonné conçu pour réaliser les tâches de gestion, collecte, stockage, analyse, flux de travail, visualisation et de réponse sur ce support. Ce système permet à une interface unifiée de gérer, cibler et accélérer des interactions dans cet espace, facilitant les relations publiques, le marketing, la publicité, le dépistage du client, le débat politique et d'autres modes de conversation orientée.
PCT/US2007/068308 2006-05-05 2007-05-04 systèmes et procédés pour une gestion de réputation de supports générés par le consommateur Ceased WO2007131213A2 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US74662106P 2006-05-05 2006-05-05
US60/746,621 2006-05-05

Publications (2)

Publication Number Publication Date
WO2007131213A2 true WO2007131213A2 (fr) 2007-11-15
WO2007131213A3 WO2007131213A3 (fr) 2008-02-14

Family

ID=38668615

Family Applications (2)

Application Number Title Priority Date Filing Date
PCT/US2007/068308 Ceased WO2007131213A2 (fr) 2006-05-05 2007-05-04 systèmes et procédés pour une gestion de réputation de supports générés par le consommateur
PCT/US2007/068392 Ceased WO2007143314A2 (fr) 2006-05-05 2007-05-07 système et procédé pour une gestion de réputation de contenus créés par le client

Family Applications After (1)

Application Number Title Priority Date Filing Date
PCT/US2007/068392 Ceased WO2007143314A2 (fr) 2006-05-05 2007-05-07 système et procédé pour une gestion de réputation de contenus créés par le client

Country Status (4)

Country Link
EP (1) EP2024814A4 (fr)
AU (1) AU2007257092B2 (fr)
CA (1) CA2690889C (fr)
WO (2) WO2007131213A2 (fr)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090287774A1 (en) * 2008-05-15 2009-11-19 Kunal Punera Method and Apparatus for Utilizing Social Network Information for Showing Reviews
EP2216743A1 (fr) * 2009-02-09 2010-08-11 Deutsche Telekom AG Procédé et serveur pour l'évaluation de la qualité d'une plate-forme de services de réseau social
US20100325107A1 (en) * 2008-02-22 2010-12-23 Christopher Kenton Systems and methods for measuring and managing distributed online conversations
CN103150432A (zh) * 2013-03-07 2013-06-12 宁波成电泰克电子信息技术发展有限公司 一种网络舆情分析方法
EP2304660A4 (fr) * 2008-06-19 2013-11-27 Wize Technologies Inc Système et procédé permettant d' agréger et de résumer un sentiment à l' égard d' un produit/d' un sujet
US9311362B1 (en) 2013-03-15 2016-04-12 Google Inc. Personal knowledge panel interface
CN109120513A (zh) * 2018-08-16 2019-01-01 湖南畅远信息技术有限公司 基于智能机器人应答的全媒体系统及实现方法
CN110209957A (zh) * 2019-06-06 2019-09-06 北京猎户星空科技有限公司 基于智能机器人的讲解方法、装置、设备及存储介质
CN119377328A (zh) * 2025-01-02 2025-01-28 中科星图数字地球合肥有限公司 一种基于消息队列的分布式时空数据库数据同步系统及方法

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010011180A1 (fr) 2008-07-25 2010-01-28 Resolvo Systems Pte Ltd Procédé et système de protection contre une divulgation de code source
US9430565B2 (en) * 2014-01-22 2016-08-30 Zefr, Inc. Providing relevant content
CN106133779B (zh) 2014-03-12 2021-03-02 南洋理工大学 电子商务企业订单接受算法控制方法和装置
CN110633312B (zh) * 2019-09-24 2022-03-25 四川长虹电器股份有限公司 一种基于Mongodb和HanLP的电商舆情分析方法及系统

Family Cites Families (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5774123A (en) * 1995-12-15 1998-06-30 Ncr Corporation Apparatus and method for enhancing navigation of an on-line multiple-resource information service
US6999938B1 (en) * 1996-06-10 2006-02-14 Libman Richard M Automated reply generation direct marketing system
US5974572A (en) * 1996-10-15 1999-10-26 Mercury Interactive Corporation Software system and methods for generating a load test using a server access log
US6826553B1 (en) 1998-12-18 2004-11-30 Knowmadic, Inc. System for providing database functions for multiple internet sources
US6691108B2 (en) 1999-12-14 2004-02-10 Nec Corporation Focused search engine and method
US20020038217A1 (en) * 2000-04-07 2002-03-28 Alan Young System and method for integrated data analysis and management
US20020062368A1 (en) * 2000-10-11 2002-05-23 David Holtzman System and method for establishing and evaluating cross community identities in electronic forums
AU2002231146A1 (en) * 2000-12-22 2002-07-08 William M. Tomlinson Collecting user responses over a network
US7644057B2 (en) * 2001-01-03 2010-01-05 International Business Machines Corporation System and method for electronic communication management
US7801909B2 (en) * 2002-10-17 2010-09-21 Poltorak Alexander I Apparatus and method for identifying and/or for analyzing potential patent infringement
US7506072B2 (en) * 2003-10-14 2009-03-17 Sun Microsystems, Inc. Web browser as web service server in interaction with business process engine
US7933958B2 (en) * 2004-06-08 2011-04-26 International Business Machines Corporation Real-time blog interaction
US7293017B2 (en) * 2004-07-01 2007-11-06 Microsoft Corporation Presentation-level content filtering for a search result
US7886024B2 (en) * 2004-07-01 2011-02-08 Microsoft Corporation Sharing media objects in a network
US7809695B2 (en) * 2004-08-23 2010-10-05 Thomson Reuters Global Resources Information retrieval systems with duplicate document detection and presentation functions
US20060080130A1 (en) * 2004-10-08 2006-04-13 Samit Choksi Method that uses enterprise application integration to provide real-time proactive post-sales and pre-sales service over SIP/SIMPLE/XMPP networks
US20060136474A1 (en) * 2004-12-21 2006-06-22 Inventec Corporation Customer service automatic answering system and method

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100325107A1 (en) * 2008-02-22 2010-12-23 Christopher Kenton Systems and methods for measuring and managing distributed online conversations
US8983975B2 (en) * 2008-02-22 2015-03-17 Christopher Kenton Systems and methods for measuring and managing distributed online conversations
US20090287774A1 (en) * 2008-05-15 2009-11-19 Kunal Punera Method and Apparatus for Utilizing Social Network Information for Showing Reviews
US8407286B2 (en) * 2008-05-15 2013-03-26 Yahoo! Inc. Method and apparatus for utilizing social network information for showing reviews
EP2304660A4 (fr) * 2008-06-19 2013-11-27 Wize Technologies Inc Système et procédé permettant d' agréger et de résumer un sentiment à l' égard d' un produit/d' un sujet
EP2216743A1 (fr) * 2009-02-09 2010-08-11 Deutsche Telekom AG Procédé et serveur pour l'évaluation de la qualité d'une plate-forme de services de réseau social
CN103150432A (zh) * 2013-03-07 2013-06-12 宁波成电泰克电子信息技术发展有限公司 一种网络舆情分析方法
US9311362B1 (en) 2013-03-15 2016-04-12 Google Inc. Personal knowledge panel interface
CN109120513A (zh) * 2018-08-16 2019-01-01 湖南畅远信息技术有限公司 基于智能机器人应答的全媒体系统及实现方法
CN110209957A (zh) * 2019-06-06 2019-09-06 北京猎户星空科技有限公司 基于智能机器人的讲解方法、装置、设备及存储介质
CN119377328A (zh) * 2025-01-02 2025-01-28 中科星图数字地球合肥有限公司 一种基于消息队列的分布式时空数据库数据同步系统及方法

Also Published As

Publication number Publication date
EP2024814A2 (fr) 2009-02-18
WO2007131213A3 (fr) 2008-02-14
WO2007143314A9 (fr) 2009-04-16
CA2690889A1 (fr) 2007-12-13
AU2007257092B2 (en) 2012-02-09
CA2690889C (fr) 2017-03-21
WO2007143314A3 (fr) 2008-10-30
AU2007257092A1 (en) 2007-12-13
WO2007143314A2 (fr) 2007-12-13
EP2024814A4 (fr) 2010-11-10

Similar Documents

Publication Publication Date Title
US7720835B2 (en) Systems and methods for consumer-generated media reputation management
US10572552B2 (en) Systems and methods for consumer-generated media reputation management
US9269068B2 (en) Systems and methods for consumer-generated media reputation management
AU2007257092B2 (en) Systems and methods for consumer-generated media reputation management
US20160217488A1 (en) Systems and methods for consumer-generated media reputation management
US20230316186A1 (en) Multi-service business platform system having entity resolution systems and methods
US9317180B2 (en) Systems and methods for consumer-generated media reputation management
US20100299326A1 (en) Apparatuses, Methods and Systems For A Forum Ferreting System
US20160063442A1 (en) Using an employee database with social media connections to calculate job candidate reputation scores
US20130031183A1 (en) Electronic mail processing and publication for shared environments
US12386797B2 (en) Multi-service business platform system having entity resolution systems and methods
US20080052147A1 (en) System and method for influencing public opinion
US20130290118A1 (en) Promoting content
US20150363803A1 (en) Business introduction interface
US10992488B2 (en) System and method for an enhanced focus group platform for a plurality of user devices in an online communication environment
WO2010065199A1 (fr) Systèmes et procédés pour une gestion de réputation de contenu généré par le consommateur
WO2009023865A1 (fr) Détermination de l'influence et du sentiment des médias générés par les consommateurs
Wandhöfer et al. Determining citizens’ opinions about stories in the news media
Wandhöfer et al. Determining citizens’ opinions about stories in the news media: analysing Google, Facebook and Twitter

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 07797348

Country of ref document: EP

Kind code of ref document: A2

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 07797348

Country of ref document: EP

Kind code of ref document: A2