WO2019000782A1 - Search term recommendation method, set top box and storage medium - Google Patents

Abstract

The invention relates to a search term recommendation method, a set-top box and an information carrier used for searching video information: acquiring an entry letter; according to said letter, in accordance with a predefined determination rule, determining relevant search terms of the letter from a predefined recommended word database; acquiring a corresponding proximity value of each of said candidate search terms; in a predefined category, to recommend candidate search words ranked from top to bottom according to the proximity value.

Description

Search term recommendation method, set top box and storage medium Technical field

The present disclosure relates to the field of information retrieval technology, for example, to a search term recommendation method, a set top box, and a storage medium.

Background technique

With the popularization of high-speed networks and the reduction of network tariffs, Internet TVs and set-top boxes have gradually entered thousands of households, providing users with rich video resources such as TV dramas and movies.

Searching as a bridge between users and video resources plays an important role in set-top boxes. In the search function, the core part is the recommendation of the search term, but the set-top box mostly uses the pinyin input combined with the local matching to recommend the search term. The advantage is that when any pinyin is input through the set top box, there will be search results, and the disadvantage is that the search result may not be what the user wants.

In addition, the existing technical solution of the search term recommendation method generally uses the video name, the popular vocabulary, and the video information such as the actor and the director as the recommended search words, and then recommends and displays it to the user. The effect of the display is not clear enough. And intuitive.

Summary of the invention

Embodiments of the present invention provide a search term recommendation method, a set top box, and a storage medium to implement accurate search term recommendation and clear presentation of search term recommendation results.

An embodiment of the present invention provides a search term recommendation method, which is applied to video information search, and includes:

Get the entered letters;

Determining, according to the predetermined letter, a candidate search term related to the letter from a preset recommendation word database according to a preset determination rule;

Obtaining a heat value corresponding to each of the candidate search words respectively;

Under the preset category, the candidate search terms are recommended in descending order of the heat value.

Optionally, according to the alphabet, according to a preset determining rule, the step of determining a candidate search term related to the letter from the preset recommendation word database includes:

Check if the letter is the correct pinyin combination;

If correct, the letters are combined as a pinyin, and the pinyin combination is searched from the preset database to obtain an index term related to the pinyin combination; or

If not correct, the letter is split into pinyin initials, and the pinyin initials are searched from the preset database to obtain index words related to the pinyin initials;

The index word is used as the candidate search word.

Optionally, the preset recommendation word database includes:

An index formed by an index word, a pinyin combination corresponding to the index word, and video information corresponding to the index word, where the index word includes a video name, a video actor name, a video director name, a video release year, and a popular word in the video. , video category or video channel name.

Optionally, the heat value includes:

Manually interpolating the tag value, using the manual intervention tag value to manually adjust the first search growth rate of the search term; and/or

a second search growth rate, the second search growth rate is equal to (Q1+Q2)/(T1-T2), wherein T1 is a current time, T2 is a creation time of the candidate search term, and Q1 is the candidate search The total number of initial cumulative searches for the word at T2, and Q2 is the total number of cumulative searches for which the candidate search term is selected for searching between T1 and T2.

Optionally, when the heat value is the manual intervention mark value and the second search growth rate, in the preset category, recommending the candidate according to the heat value from high to low The steps to search for words include:

In the preset category, the candidate search terms are first recommended in the order of the second search growth rate according to the order of the manual intervention mark values from high to low.

Optionally, after the step of recommending the candidate search term in the order of the heat value from high to low, in the preset category, the method further includes:

After the candidate search term is selected for searching, the Q2 value of the candidate search term is updated.

Optionally, the number of the candidate search words is 20.

Optionally, the preset category includes: a video actor name and/or a video director name.

An embodiment of the present invention provides a set top box, where the set top box includes:

One or more first processors;

a first storage device configured to store one or more programs;

When the one or more programs are executed by the one or more first processors, the one or more first processors implement a search term recommendation method as described in any of the embodiments of the present invention.

Embodiments of the present invention provide a computer readable storage medium storing computer executable instructions that are executed by a computer processor to implement the search term recommendation method described above.

The embodiment of the invention realizes a clearer recommendation word display effect, ensures the accuracy of the search word recommendation, and is convenient for the user to quickly and conveniently find the required search result according to the recommendation word.

DRAWINGS

FIG. 1 is a schematic diagram of an application scenario provided by an embodiment;

2 is a flowchart of a search term recommendation method according to an embodiment;

FIG. 3 is a flowchart of acquiring video play amount information according to an embodiment; FIG.

4 is a flowchart of obtaining director and actor heat information provided by an embodiment;

FIG. 5 is a schematic diagram of a search box of a set top box according to an embodiment; FIG.

6 is a flowchart of a search term recommendation method provided by another embodiment;

FIG. 7 is a schematic structural diagram of a set top box according to an embodiment.

Detailed ways

FIG. 1 is an application scenario provided by an embodiment. As shown in FIG. 1 , a smart terminal (such as a smart set top box, a portable smart terminal such as a mobile phone or a tablet computer, or a media play box, etc., here, the smart set top box 110 is taken as an example) can be connected with a remote controller. 111. The television 112 and the cloud video resource library 113 respectively interact, that is, the smart set top box 110 is communicably connected to the remote controller 111, the television 112, and the cloud video resource library 113, respectively.

When the user wants to watch the video, it is necessary to input letters in the input box of the search interface of the smart set top box 110 by using the remote controller 111 or the virtual keyboard of the television set 112, and then the recommendation box on the search interface of the smart set top box 110 will be Given the recommended search words corresponding to the input letters, the user can select the video resources on the cloud video resource library 113 to complete the corresponding video playback as long as the user selects according to the requirements.

2 is a flowchart of a search term recommendation method provided by an embodiment, which is applicable to a scenario of searching for video information by using pinyin or alphabet. The method is performed by a set top box, and the set top box can be implemented by software and/or hardware. Including the following steps:

In step 210, the entered letters are obtained.

In the embodiment of the present invention, the manner of obtaining is extracted from the input box when the input information is confirmed to be correct, and the input letter is a full name or abbreviation of pinyin or English.

The user connects the smart set top box 110 through the remote controller 111, inputs the letters to the input box of the search interface of the smart set top box 110 through the numeric keypad on the remote controller 111, and displays it through the television 112. The smart set top box 110 obtains input letters in the input box to implement determination of subsequent search terms.

When the user inputs a letter in the input box and wants to search for the video information through the letter, in order to facilitate the user's search, the function of the search word recommendation is often set in the lower part of the input box or in the specific recommendation box area, It is convenient for users to accurately select search terms and reduce user input costs.

Optionally, the input manner of the letter may be keyboard input, handwriting input or voice input, and the input letter may be a pinyin combination or a pinyin initial letter. In order to avoid that the recommended search word does not meet the user's situation, generally, the requirement is required. The letters entered are complete and correct.

In step 220, according to the letter, a candidate search term related to the letter is determined from the preset recommendation word database according to a preset determination rule.

Optionally, the letter may match a plurality of candidate search words, and if all the matches of the letters are determined as much as possible, a certain rule may be followed. The preset determination rule in the embodiment of the present invention is to check whether the letter is a correct pinyin combination, and then perform a corresponding operation according to the check result of the letter. Alternatively, the letter may be directly used as a pinyin combination. The determination of the candidate search term is performed or the letter is split into pinyin initials for determination of the candidate search term. The determined candidate search term must be associated with the letter.

Optionally, the preset recommendation word database includes:

An index formed by an index word, a pinyin combination corresponding to the index word, and video information corresponding to the index word, where the index word includes a video name, a video actor name, a video director name, a video release year, and a popular word in the video. , video category or video channel name.

Optionally, Table 1 is an index table. Referring to Table 1, the index table includes: an index word, a pinyin combination corresponding to the index word, and video information corresponding to the index word. Through the index table, the information to be queried can be retrieved from the preset recommendation word database more easily.

The index table is established in the initial video resource integration process, and the video data information (such as video name, video actor name, etc.) in the integrated video resource is extracted by the search system, and converted into pinyin and stored in the preset. Recommended in the database.

In step 230, the heat values corresponding to the candidate search words are respectively acquired.

Optionally, after determining the candidate search term related to the letter from the preset recommendation word database according to the alphabet, the determined candidate search words need to be sorted first. The candidate search words are recommended to the user, and the manner of sorting is determined according to the respective heat values of the candidate search words.

Optionally, the heat value includes:

Manually interpolating the tag value, using the manual intervention tag value to manually adjust the first search growth rate of the search term; and/or

a second search growth rate, the second search growth rate is equal to (Q1+Q2)/(T1-T2), wherein T1 is a current time, T2 is a creation time of the candidate search term, and Q1 is the candidate search The total number of initial cumulative searches for the word at T2, and Q2 is the total number of cumulative searches for which the candidate search term is selected for searching between T1 and T2.

Optionally, the popularity value of the search term is a cumulative process. Optionally, it may be all the heat statistics of the search term from the beginning to the present, or all the heat statistics of the cycle of the month or quarter. The heat value is divided into two parts, one part is a manual intervention mark value, the addition and deletion of the manual intervention mark value determines the size of the first search growth rate, and the other part is the second search growth rate. The manual intervention tag value refers to the background search worker in the cloud adjusting the first search growth rate of one or some search words by manual intervention, and the first search growth rate is controlled by humans, not normal growth data. Statistics, the purpose is to make some or some of the sudden fires but the total growth rate has not been able to rank The search term in the forefront can be recommended to users in this era of information explosion.

The second search growth rate is a statistic of normal growth data of the search term. In the embodiment of the present invention, the manner in which the search term is a part of the initial heat value of the video name is obtained, as shown in FIG. 3, optionally, by acquiring the video play quantity information of the plurality of partner video sources, and determining The search term is the video actor name, the video director name part of the initial heat information acquisition method, please refer to Figure 4, specifically through the Baidu Billboard or 360 hot list in real time.

The popularity value of the search term is composed of the manual intervention marker value and the second search growth rate, and the manual intervention marker value has a higher priority than the second search growth rate, that is, the search is determined When the heat value of the word is high or low, the manual intervention mark value is preferentially referred to and then referred to the second search growth rate.

In step 240, under the preset category, the candidate search terms are recommended in descending order of the heat value.

In the embodiment of the present invention, the preset category refers to the category of the search term corresponding to the letter input by the user. On the basis of the above technical solutions, the embodiment of the present invention enriches the presentation form of the recommended search words by adding a preset category, and the effect of the display is clear and intuitive.

The preset categories include: video name, video actor, and/or video director. The relationship between the preset categories may be a parallel relationship, a merge relationship, or a prioritized relationship.

Optionally, for example, inputting the letter huojianhua in the input box, the matching search word is Huo Jianhua, which automatically matches the video actor in the preset category; inputting the letter renmingdemingyi in the input box, can match the search word to the name of the people, and its automatic matching To the name of the video in the preset category; enter the letter zhouxingchi in the input box to match the search term to Stephen Chow, which automatically matches the video actor and video director in the preset category. There is a mapping table between the preset category and the search term, and the mapping is performed through the mapping table. The preset category and the search term mapping table may be one or more. When the preset category and the search term mapping table are plural, the priority levels of the plurality of mapping tables may be set.

Under the preset category, the candidate search words are recommended in descending order of the heat value, which enables the user to find more convenient and quick, and can quickly find the favorite video name, video actor and/or video director. And other information.

In general, the preset category is not classified before the input letter is obtained, and the system intelligent default preset category; in other embodiments, the input letter may be obtained according to the preset category before the input letter is acquired.

Optionally, the search terms recommended under the preset category are provided to the user in the form of a pull-down list in descending order of the heat value.

Optionally, the number of the candidate search words recommended is 20.

Optionally, the recommended search term is the top N search term of the popularity value, where N is configurable, typically 20.

In the embodiment of the present invention, the entire process of implementing search term recommendation by inputting a letter is performed on the search interface, and is presented by the television 112. The search interface is as shown in FIG. 5.

In the embodiment of the present invention, the first technical means for determining the corresponding candidate search term from the preset recommendation word database by using the preset determination rule is adopted, and the search term determination is improved relative to the prior art. Accuracy; then by obtaining the heat value of the candidate search words, the second technical means of sorting under the preset category improves the accuracy of the search term recommendation and achieves a clear and intuitive result presentation effect. That is, by obtaining the input letter; according to the letter, according to the preset determination rule, determining, from the preset recommendation word database, the candidate search words related to the letter; respectively acquiring the heat value corresponding to each of the candidate search words; Under the preset category, according to the order of the heat value from high to low, the technical solution of the candidate search term is recommended, and compared with the existing search word recommendation technology, a clearer recommendation word display effect can be realized, and the search is guaranteed. The accuracy of the word recommendation is convenient for the user to quickly and conveniently find the desired search result according to the recommended word.

Optionally, after the step of recommending the candidate search term in the preset category, the method further includes:

After the candidate search term is selected for searching, the Q2 value of the candidate search term is updated.

Optionally, the more frequently the user selects the candidate search words for the search, the more accurate the order of the recommended search words is, which allows the user to optimize the recommendation of the search words without knowing it in the actual use process. Accuracy.

FIG. 6 is a method for recommending a search term according to an embodiment. On the basis of the foregoing embodiment, step 220 “determines a rule related to the letter from a preset recommendation word database according to the preset determination rule according to the letter. Further optimization of candidate search terms. include:

Check if the letter is the correct pinyin combination;

If correct, the letters are combined as a pinyin, and the pinyin combination is searched from the preset database to obtain an index term related to the pinyin combination; or

If not correct, the letter is split into pinyin initials, and the pinyin initials are searched from the preset database to obtain index words related to the pinyin initials;

The index word is used as the candidate search word.

As shown in FIG. 6, the search term recommendation method provided by an embodiment includes the following steps:

In step 610, the entered letters are obtained.

In step 620, it is checked whether the letter is the correct pinyin combination. If yes, step 630 is performed, and if not, step 640 is performed.

In this embodiment, after the smart set top box 110 acquires the letter input by the remote controller 111 in the preset category, the letter is checked, specifically, whether the letter is a correct pinyin combination, or alternatively, by checking the Whether the pinyin combination of the letters is determined by the matching of the pinyin combinations corresponding to the index words in the preset recommendation word database, or may be determined by checking whether the pinyin combination composed of the letters can obtain Chinese characters or English words; In the latter manner, the input of the letters needs to be matched with a specific input method, so that it is convenient to quickly confirm whether the letters can form a correct pinyin combination. Optionally, after inputting a number of letters of a pinyin combination, a plurality of letters combined with the next pinyin are distinguished by a space.

In step 630, the letters are combined as pinyin, and the pinyin combination is searched from the preset database to obtain index words related to the pinyin combination.

In step 640, the letter is split into pinyin initials, and the pinyin initials are searched from the preset database to obtain index words related to the pinyin initials.

The related index words obtained by searching the letters as pinyin combinations or pinyin initials may be one or more, and each index word has corresponding video information.

In step 650, the index word is used as the candidate search term.

The index word, the pinyin combination corresponding to the index word, and the video information corresponding to the index word have a one-to-one correspondence. By matching the pinyin combination of the letters with the pinyin combination in the recommendation word database, the corresponding index word, even the video information corresponding to the index word, can be determined.

In step 660, the heat values corresponding to the candidate search words are respectively acquired.

In step 670, under the preset category, the candidate search terms are recommended in descending order of the heat value.

Optionally, under the preset category, recommending the candidate search term according to the order of the heat value from high to low, including:

In the preset category, the candidate search terms are first recommended in the order of the second search growth rate according to the order of the manual intervention mark values from high to low.

Optionally, under the preset category, the candidate search terms are recommended in the form of a pull-down list according to the order of the second search growth rate in the order of the manual intervention mark values from high to low.

In the embodiment of the present invention, by inputting the letter that needs to be queried, and then checking the correctness of the pinyin combination of the letter, the search word is determined as the pinyin combination or the pinyin initial, and then the recommended search word is displayed, which can be realized. The clear recommendation word display effect ensures the accuracy of the search term recommendation, so that the user can quickly and conveniently find the required search result according to the recommendation word.

This embodiment provides a set top box based on the above embodiment. As shown in FIG. 7, the set top box 712 is represented in the form of a universal set top box device. The components of the set top box 712 can include, but are not limited to, one or more first processors or first processing units 716, system memory 728, and a bus 718 that connects different system components, including system memory 728 and first processing unit 716.

Optionally, when one or more programs included in the set top box 712 are executed by the one or more first processors or the first processing unit 716, the program performs the following operations:

Get the entered letters;

Determining, according to the predetermined letter, a candidate search term related to the letter from a preset recommendation word database according to a preset determination rule;

Obtaining a heat value corresponding to each of the candidate search words respectively;

Under the preset category, the candidate search terms are recommended in descending order of the heat value.

Bus 718 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a first processor, or a local bus using any of a variety of bus structures. Optionally, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA Bus, a Video Electronics Standards Association (VESA) local bus, and peripheral component interconnects ( PCI) bus.

Set top box 712 typically includes a variety of computer system readable media. These media can be anything that can Available media accessed by set top box 712 include volatile and non-volatile media, removable and non-removable media.

System memory 728 can include, but is not limited to, computer system readable media in the form of volatile memory, such as random access memory (RAM) 730 and/or cache memory 732. Set top box 712 may further include other removable/non-removable, volatile/non-volatile computer system storage media. By way of example only, storage system 734 can be configured to read and write non-removable, non-volatile magnetic media (not shown in Figure 7, commonly referred to as a "hard disk drive"). Disk drives that can read and write to removable non-volatile disks (such as "floppy disks"), as well as read and write to removable non-volatile discs (such as CD-ROMs, DVD-ROMs, or other optical media) Optical drive. In these cases, each drive can be coupled to bus 718 via one or more data medium interfaces. Memory 728 can include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of various embodiments of the present invention.

Program/utility 740 having a set (at least one) of program modules 742, which may be stored, for example, in memory 728, such program program 742 includes, but is not limited to, an operating system, one or more applications, other program modules, and program data. An implementation of the network environment may be included in each or some of these examples. Program module 742 typically performs the functions and/or methods of the described embodiments of the present invention.

The set top box 712 can also be in communication with one or more external devices 717 (e.g., pointing device, display 724, etc.), and can also be in communication with one or more devices that enable a user to interact with the set top box 712, and/or with the set top box 712. Any device (eg, a network card, modem, etc.) capable of communicating with one or more other computing devices communicates. This communication can take place via an input/output (I/O) interface 722. Also, set top box 712 can also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and/or a public network, such as the Internet) via network adapter 720. As shown, network adapter 720 communicates with other modules of set top box 712 via bus 718. It should be understood that although not shown in the figures, other hardware and/or software modules may be utilized in connection with set top box 712, including but not limited to: microcode, device drivers, redundant first processing units, external disk drive arrays, RAID systems, tapes. Drives and data backup storage systems, etc.

The first processing unit 716 performs various function applications and data processing by running a program stored in the system memory 728, for example, implementing a search term recommendation method provided by an embodiment of the present invention.

The embodiment of the present invention further provides a computer readable storage medium storing computer executable instructions, which when executed by a computer processor, is configured to perform a search term recommendation method, including:

Get the entered letters;

Determining, according to the predetermined letter, a candidate search term related to the letter from a preset recommendation word database according to a preset determination rule;

Obtaining a heat value corresponding to each of the candidate search words respectively;

Under the preset category, the candidate search terms are recommended in descending order of the heat value.

Optionally, a storage medium containing computer executable instructions provided by the embodiment of the present invention is not limited to the foregoing method, and may also perform a search term recommendation method provided by any embodiment of the present invention.

The invention can be implemented by means of software and necessary general hardware, and of course can also be implemented by hardware. Based on such understanding, the embodiments of the present invention may be embodied in the form of a software product, which may be stored in a computer readable storage medium, such as a computer floppy disk, read-only memory (ROM), random. Access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform the embodiments of the present invention. The method described.

Optionally, in the embodiment of the set top box, the included units and modules are only divided according to functional logic, but are not limited to the foregoing division, as long as the corresponding functions can be implemented; The specific names are also for convenience of distinguishing from each other and are not intended to limit the scope of the present invention.

Industrial applicability

The embodiment of the invention realizes a clearer recommendation word display effect, ensures the accuracy of the search word recommendation, and is convenient for the user to quickly and conveniently find the required search result according to the recommendation word.

Claims (10)

A search term recommendation method applied to video information search, including: Get the entered letters; Determining, according to the predetermined letter, a candidate search term related to the letter from a preset recommendation word database according to a preset determination rule; Obtaining a heat value corresponding to each of the candidate search words respectively; Under the preset category, the candidate search terms are recommended in descending order of the heat value. The method according to claim 1, wherein the step of determining a candidate search term related to the letter from the preset recommendation word database according to the letter according to a preset determination rule comprises: Check if the letter is the correct pinyin combination; If correct, the letters are combined as a pinyin, and the pinyin combination is searched from the preset database to obtain an index term related to the pinyin combination; or If not correct, the letter is split into pinyin initials, and the pinyin initials are searched from the preset database to obtain index words related to the pinyin initials; The index word is used as the candidate search word. The method of claim 1, wherein the predetermined recommendation word database comprises: An index formed by an index word, a pinyin combination corresponding to the index word, and video information corresponding to the index word, where the index word includes a video name, a video actor name, a video director name, a video release year, and a popular word in the video. , video category or video channel name. The method of claim 1 wherein said heat value comprises: Manually interpolating the tag value, using the manual intervention tag value to manually adjust the first search growth rate of the search term; and/or a second search growth rate, the second search growth rate is equal to (Q1+Q2)/(T1-T2), wherein T1 is a current time, T2 is a creation time of the candidate search term, and Q1 is the candidate search The total number of initial cumulative searches for the word at T2, and Q2 is the total number of cumulative searches for which the candidate search term is selected for searching between T1 and T2. The method according to claim 4, wherein when said heat value is said manual intervention flag value and said second search growth rate, said predetermined value is in accordance with said heat value from high to low The order of recommending the candidate search terms includes: In the preset category, the candidate search terms are first recommended in the order of the second search growth rate according to the order of the manual intervention mark values from high to low. The method according to claim 5, wherein, after the step of recommending the candidate search term in the order of the heat value from high to low, in the preset category, the method further comprises: After the candidate search term is selected for searching, the Q2 value of the candidate search term is updated. The method of claim 1 wherein the number of candidate search terms is 20. The method of claim 1 wherein the predetermined category comprises: a video actor name and/or a video director name. A set top box comprising: One or more first processors; a first storage device configured to store one or more programs; When the one or more programs are executed by the one or more first processors, the one or more first processors implement the search term recommendation method according to any one of claims 1-8. A storage medium comprising computer executable instructions, the computer executable instructions being executed by a computer processor to implement a search term recommendation method according to any of claims 1-8.

Images