CN115858603A - An intelligent retrieval method for business start-up standard matters that integrates multiple attributes - Google Patents

Abstract

This application discloses an intelligent search method for standard items of business start-up that integrates multiple attributes. The method includes steps: including steps: S1, preprocessing and statistical analysis of enterprise historical registration data, and obtaining the occurrence times of each industry classification; S2, business start-up The catalog of standard items is generated; S3, the multi-attribute circle construction of standard items for business start-up, forming a complete multi-attribute knowledge base for business start-up items; S4: the output of intelligent retrieval results based on the multi-attribute knowledge base for business start-up items. This application can not only solve the problem of the current incomplete catalog of business start-up items, but also solve the problem that the service provider is unable to start with a large number of optional business start-up items. At the same time, it can also solve the business scope recommendation and business start-up. Two problems are well solved, from the multiple search terms that service workers may input to the precise positioning of pending business start-up items, and provide more convenient ways for ordinary people to do business.

Description

An intelligent retrieval method for business start-up standard matters that integrates multiple attributes

technical field

This application relates to the technical field of information processing, in particular, to an intelligent retrieval method for business start-up standard items that integrates multiple attributes.

Background technique

In recent years, the State Administration of Market Regulation has required that one thing be done once for starting a business, and measures should be taken to reduce the time for starting a business to within 4 working days. Local governments have begun to implement online business start-up platforms, which will affect the common people and staff Higher requirements were put forward. In the past, when handling offline, the operator and the staff can communicate face-to-face on the specific content of starting a business, but when handling online, it is necessary for the operator to have a certain understanding of the specific content of starting a business in advance, and an online business starting platform is also required Provides a complete alternative guide catalog for business start-up matters, but most of the current business start-up matters catalogs are incomplete. Choosing the correct name of the business start-up item is the first step to quickly and accurately complete the business registration. However, in reality, due to the deviation between the operator's own understanding of the to-do item and the standard understanding of the to-do item, there will often be a business There is no obvious relationship between search results and the search keywords entered by the service provider.

Contents of the invention

In view of the above technical problems, on the one hand, the present application provides an intelligent retrieval method for business start-up standard items that integrates multiple attributes.

The technical scheme that this application adopts is as follows:

An intelligent retrieval method for business start-up standard matters that integrates multiple attributes, comprising the steps of:

S1. Preprocessing and statistical analysis of enterprise historical registration data: read individual and enterprise historical registration data, clean and preprocess the data, and then conduct statistical analysis based on the processed data to obtain the occurrence times of each industry classification;

S2. Generation of the catalog of standard items for starting a business: Based on the classification of national economic industries, combined with the historical registration data of individuals and companies, a catalog of standard items for starting a business is generated;

S3. Construction of multi-attribute circles of standard items for business start-up: define the concept of multi-attribute circles for standard items of business start-up, obtain relevant knowledge of business start-up through natural language processing technology for each circle, and form a complete multi-attribute knowledge base of business start-up items ;

S4: Output of intelligent retrieval results based on multi-attribute knowledge base of business start-up items: With the help of natural language processing technology and synonym model, obtain the keyword list of retrieval keywords, and determine the matching score of different items according to the attribute type hit by keywords, so as to Provide intelligent search result recommendations for business start-up matters.

Further, the step S1 specifically includes the steps of:

Step S11, missing value processing: read the historical registration data of individuals and enterprises, and delete the data that lacks the name of the trade name in the data, that is, the data that lacks the name of the enterprise or the name of the individual;

Step S12. Group statistical analysis based on industry classification: Based on the industry classification field in the original data, perform group statistics on the data to obtain the occurrence times of each industry classification.

Further, the step S2 specifically includes the steps of:

Step S21, Preliminarily forming a catalog of standard items for starting a business based on national standards: First, based on the subdivision of industries in the National Economic Industry Classification Table, combined with the specific connotation of the industry classification and business knowledge related to starting a business, merge to obtain a preliminary version of the catalog of standard items for starting a business;

Step S22. Adjusting the catalog of standard items for starting a business based on regional characteristics: In order to reflect the spatial and temporal characteristics of starting a business in a region, adjust the preliminary version of the catalog of standard items obtained in step S21 in combination with the statistical analysis results of historical registration data of individuals and companies in the region within one year , forming a complete catalog of standard items for starting a business.

Further, the step S3 specifically includes:

Step S31: Expansion of the industry form circle: First, use the content of the business scope entry to expand the industry form, that is, according to the business scope registration specification to express the corresponding relationship between the business scope entry and the industry classification in the catalog, and match the business scope entries corresponding to each industry classification According to the content of the activity, the industry form is obtained; combined with the corresponding relationship between the industry code and the standard items for starting a business, the industry form is associated with the standard items for starting a business, so as to realize the expansion of the industry form circle corresponding to each standard item for starting a business;

Step S32. Expansion of the event expression circle: Based on the historical registration data of individuals and enterprises, word segmentation is performed on the registered names of individuals and enterprises to obtain commonly used industry expressions, that is, colloquial expressions in enterprise registered names; combined with industry codes and enterprise start-up Corresponding relationship of standard items, linking industry expressions with standard items for starting a business, so as to realize the expansion of the industry category circle corresponding to each standard item for starting a business;

Step S33. Expansion of the brand circle: Combining crawler and natural language processing technology, referring to the content of websites including China Brand Network, China Franchise Network, and Maigou.com to obtain brand data of various industries, and compare brand data with business start-up standards Correlation to form a complete multi-attribute knowledge base of business start-up matters.

Further, step S33 specifically includes the following steps:

S331. Acquisition of brand data: obtain brand data from China Brand Network, China Franchise Network, and Maigou Network, and perform structured processing on the data;

S332. Matching of brand data and business start-up items: use natural language processing technology to realize automatic matching of brand data and business start-up items, and manually verify the matched data and associate unmatched data, thereby completing the brand to business start-up standard The mapping of items forms a complete multi-attribute knowledge base of business start-up items.

Further, the step S4 specifically includes the steps of:

S41. Firstly, perform a full-text matching query of the input search keyword and the multi-attribute knowledge base of business start-up matters, the industry form, item expression and brand words, and if they can be completely matched, directly recommend the business start-up standard items associated with the keyword;

S42. If no exact matching word is found, use natural language processing technology to perform word segmentation processing on the search keyword to obtain a search keyword group, use the error correction model to obtain the correct input search keyword, and then obtain the search keyword according to the synonym model synonymous phrases with similar words;

S43. Use the result of step S42 as the search keyword to search the standard items for starting a business and its multi-circle attributes, and return the one with the highest score according to the matching degree of the search keyword and the set search keyword and attribute field weight order N items of standard items for starting a business are recommended as intelligent search results.

Further, the step S43 specifically includes the steps of:

S431. When retrieving the catalog of standard matters for starting an enterprise, perform weight distribution, wherein, the weight Wt _i of the retrieval keyword is distributed as follows: the weight of the original retrieval keyword is Wt ₁ , and the word segmentation keyword obtained by the word segmentation algorithm in step S42 The weight of the phrase is Wt ₂ and the weight of the synonymous keyword phrase is Wt ₃ obtained according to the synonym model, and the weight of the retrieval keyword is set to Wt ₁ >Wt ₂ >Wt ₃ during retrieval; among them, for the search object enterprise to open a catalog of standard items and The distribution principle of the weight Ws _j of its multiple attributes is as follows: the weight of the standard item name obtained in step S2 is Ws ₁ , the weight of the industry form attribute obtained in step S31 is Ws ₂ , the weight of the item expression attribute obtained in step S32 is Ws ₃ , and step S33 The weight of the obtained brand attribute is Ws ₄ , and the weight of the retrieval target attribute is set to Ws ₁ >Ws ₂ >Ws ₃ >Ws ₄ during retrieval;

S432, the score calculation algorithm of the search term hitting the standard item is traditional TF-IDF, and the score of the search term hitting the standard item is Sij, wherein, i represents the source of the search term (i=1 indicates that the search term is an input search keyword , i=2 indicates that the search term is the word segmentation result of the input search keyword, i=3 indicates that the search term is a synonym obtained according to the synonym model); j indicates the attribute circle layer hit by the search term (j=1 indicates that the search term What the term hits is the standard item name, j=2 means that what the search term hits is the industry form attribute, j=3 means that what the search term hits is the item expression attribute, and j=4 means that what the search term hits is the brand attribute), Among them, _S11 represents the score of the input retrieval keyword hitting the name of the standard item; _S21 represents the score of the word segmentation keyword hitting the name of the standard item; _S32 represents the score of the synonymous keyword hitting the industry form attribute; _S34 represents the synonymous key The score of the word hitting the brand attribute;

Compute the final score for the criteria item being retrieved hits:

Among them, Wt _i is the weight of the search keywords, Ws _j is the weight of the catalog of standard items for starting a business and its multi-attributes, S _ij is the score of the search terms hitting the standard items, sorted and returned according to the final scores of each standard item retrieved The N items with the highest scores as the standard items for starting a business will be used as the recommended results of the intelligent search.

Another aspect of the present application also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and operable on the processor. When the processor executes the program, the fusion of multiple attributes is realized. The steps of the intelligent retrieval method for the standard matters for the establishment of the enterprise.

Another aspect of the present application also provides a storage medium, the storage medium includes a stored program, and when the program is running, the device where the storage medium is located is controlled to execute the intelligent retrieval method for enterprise start-up standard items that integrate multiple attributes A step of.

Compared with the prior art, the present application has the following beneficial effects:

This application provides an intelligent retrieval method for business start-up standard items that integrates multiple attributes. The method includes steps: S1. Preprocessing and statistical analysis of historical registration data of enterprises: reading individual and historical registration data of enterprises, cleaning and preprocessing the data , and then conduct statistical analysis based on the processed data to obtain the occurrence times of each industry classification; S2. Generation of the catalog of standard items for starting a business: based on the classification of national economic industries, combined with the historical registration data of individuals and companies, a catalog of standard items for starting a business is generated; S3. Construction of multi-attribute circles of standard items for business start-up: define the concept of multi-attribute circles for standard items of business start-up, obtain relevant knowledge of business start-up through natural language processing technology for each circle, and form a complete multi-attribute knowledge base of business start-up items ; S4: Output of intelligent retrieval results based on multi-attribute knowledge base of business start-up items: with the help of natural language processing technology and synonym model, obtain the keyword list of retrieval keywords, and determine the matching scores of different items according to the attribute types hit by keywords, In this way, the intelligent retrieval result recommendation of the business start-up items is given.

This application starts from the industry classification of the national standard, combined with the historical data of local industrial and commercial registration, can establish a complete list of standard items for business start-up with regional characteristics, and combined with natural language processing technology, it realizes the expansion and improvement of multiple attributes of business start-up items. Therefore, the multi-attribute knowledge base of business start-up items is constructed, and finally an intelligent retrieval and recommendation method for business start-up items is proposed based on the multi-attribute knowledge base. In this way, it is possible to quickly and accurately provide service providers with intelligent retrieval results of standard items for starting a business.

This application mainly solves the following technical problems: Combining national industry classification standards and local industrial and commercial registration historical data, a complete catalog of business start-up standard items is established; the concept of multiple attribute circles is defined for items, and each circle is naturally The language processing technology obtains relevant expanded thesaurus; based on the multi-attribute circle of business start-up matters, intelligent retrieval technology is used to realize all-round rapid and accurate positioning from search words to matters. Based on this application, it can not only solve the problem of the current incomplete catalog of business start-up items, but also solve the problem that the service provider is unable to start with a large number of optional business start-up items. At the same time, it can also solve the business scope recommendation and business start-up requirements. Handling two problems at a time can well solve the problems ranging from the multiple search terms that service workers may input to the precise positioning of to-do business start-up items, and provide more convenient ways for ordinary people to do business.

In addition to the objects, features and advantages described above, the present application has other objects, features and advantages. The application will be described in further detail below with reference to the accompanying drawings.

Description of drawings

The drawings constituting a part of the application are used to provide further understanding of the application, and the schematic embodiments and descriptions of the application are used to explain the application, and do not constitute an improper limitation to the application. In the attached picture:

Fig. 1 is a schematic flowchart of an intelligent retrieval method for business start-up standard items integrating multiple attributes in a preferred embodiment of the present application.

Fig. 2 is a schematic flowchart of the sub-steps of step S1 in the preferred embodiment of the present application.

Fig. 3 is a schematic flowchart of the sub-steps of step S2 in the preferred embodiment of the present application.

Fig. 4 is a schematic flowchart of the sub-steps of step S3 in the preferred embodiment of the present application.

Fig. 5 is a schematic flowchart of the sub-steps of step S33 in the preferred embodiment of the present application.

Fig. 6 is a schematic flowchart of the sub-steps of step S4 in the preferred embodiment of the present application.

Fig. 7 is a schematic block diagram of an electronic device entity in a preferred embodiment of the present application.

FIG. 8 is an internal structure diagram of a computer device in a preferred embodiment of the present application.

Detailed ways

It should be noted that, in the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and embodiments.

Referring to Figure 1, in one embodiment of the present invention, an intelligent retrieval method for business start-up standard items is provided that integrates multiple attributes, based on the national economic industry classification specification document implemented on October 1, 2017, combined with the use of XX Province 2021.10 .15 to 2022.10.15, according to the process shown in Figure 1, first, preprocess and statistically analyze the historical registration data of enterprises, and then sort out the business start-up items according to the national economic industry classification specification documents The preliminary catalog is adjusted according to the regional characteristics reflected in the historical data, and finally 693 business start-up items are obtained to form a standard catalog; secondly, a knowledge base of business start-up items is generated by using crawler technology and natural language processing technology; finally, the above-mentioned The knowledge base of business start-up matters is synchronized to the ES database, and based on the multi-attribute knowledge base of business start-up matters, intelligent retrieval and recommendation of standard business start-up matters are realized, including steps:

S1. Preprocessing and statistical analysis of enterprise historical registration data: read individual and enterprise historical registration data, clean and preprocess the data, and then conduct statistical analysis based on the processed data to obtain the occurrence times of each industry classification;

S2. Generation of the catalog of standard items for starting a business: Based on the classification of national economic industries, combined with the historical registration data of individuals and companies, a catalog of standard items for starting a business is generated;

S3. Construction of multi-attribute circles of standard items for business start-up: define the concept of multi-attribute circles for standard items of business start-up, obtain relevant knowledge of business start-up through natural language processing technology for each circle, and form a complete multi-attribute knowledge base of business start-up items ;

S4: Output of intelligent retrieval results based on multi-attribute knowledge base of business start-up items: With the help of natural language processing technology and synonym model, obtain the keyword list of retrieval keywords, and determine the matching score of different items according to the attribute type hit by keywords, so as to Provide intelligent search result recommendations for business start-up matters.

In this embodiment, starting from the industry classification of national standards, combined with the historical data of local industrial and commercial registration, a complete catalog of business start-up standard items with regional characteristics can be established, and combined with natural language processing technology, the expansion and improvement of multiple attributes of business start-up items are realized. , so as to build a multi-attribute knowledge base of business start-up items, and finally propose an intelligent retrieval and recommendation method for business start-up items based on the multi-attribute knowledge base. In this way, it is possible to quickly and accurately provide service providers with intelligent retrieval results of standard items for starting a business.

This embodiment mainly solves the following technical problems: Combining the national industry classification standards and local industrial and commercial registration historical data, a complete catalog of business start-up standard items is established; The natural language processing technology obtains relevant expanded thesaurus; based on the multi-attribute circle of business start-up matters, intelligent retrieval technology is used to realize all-round rapid and accurate positioning from search terms to matters. Based on this embodiment, it can not only solve the problem that the current list of business start-up items is incomplete, but also solve the problem that the service provider is unable to start with a large number of optional business start-up items. At the same time, it can also solve the business scope recommendation and business start-up requirements in the future. The problem of handling two items at a time is a good solution to the problem of multiple search terms that service providers may input to the precise positioning of pending business start-up items, and provides a more convenient way for ordinary people to handle affairs.

In one embodiment of the present invention, based on the national economic industry classification specification document implemented on October 1, 2017, combined with the individual and enterprise registration data of XX province from 2021.10.15 to 2022.10.15, as shown in Figure 1 Firstly, preprocessing and statistical analysis are performed on the historical registration data of the enterprise, and then a preliminary catalog of business start-up items is sorted out according to the national economic industry classification specification documents, and the preliminary catalog is adjusted in combination with the regional characteristics reflected in the historical data, and finally 693 Form a standard catalog of business start-up items. Secondly, use crawler technology and natural language processing technology to generate a knowledge base of business start-up matters. Finally, the above-mentioned enterprise start-up item knowledge base is synchronized to the ES database, and based on the enterprise start-up item multi-attribute knowledge base, the intelligent retrieval and recommendation of enterprise start-up standard items is realized.

Preferably, as shown in Figure 2, the step S1 specifically includes the steps of:

Step S11. Missing value processing: read the historical registration data of individuals and enterprises, and delete the data that lacks the name of the trade name in the data, that is, the data that lacks the name of the enterprise or the name of the individual. This embodiment reads XX province from 2021.10.15 to 2022.10.15 Individual and enterprise registration data within one year, including 1,191,164 individual industrial and commercial household registration data, and 356,139 enterprise registration data, 212,984 individual industrial and commercial household registration data missing the name of the company after reading, and enterprise registration data No data lacking font size and name was found, so the remaining 1,334,319 pieces of data were finally processed;

Step S12. Group statistical analysis based on industry classification: Based on the industry classification field in the original data, perform group statistics on the data to obtain individual or enterprise registration frequencies under each industry classification.

Preferably, as shown in Figure 3, the step S2 specifically includes the steps of:

Step S21. Preliminarily form a catalog of standard items for starting a business based on national standards: firstly, based on the subdivision of industries in the National Economic Industry Classification Table, combined with the specific connotation of the industry classification and business knowledge related to starting a business, merge and obtain 600 standard items for starting a business to form a business start The preliminary version of the catalog of standard items;

Step S22. Adjusting the catalog of standard items for starting a business based on regional characteristics: In order to reflect the spatial and temporal characteristics of starting a business in a region, adjust the preliminary version of the catalog of standard items obtained in step S21 in combination with the statistical analysis results of historical registration data of individuals and companies in the region within one year , to form a complete list of standard items for starting a business. In this embodiment, according to the statistical results obtained in step S3, the classification of national economic industries whose frequency of occurrence is greater than or equal to one-thousandth of the total data volume (that is, the frequency of occurrence ≥ 1334) is separately regarded as a standard item. , and finally got 693 standard items for starting a business.

Preferably, as shown in Figure 4, the step S3 specifically includes:

Step S31: Expansion of the industry form circle: First, use the content of the business scope entry to expand the industry form, that is, according to the business scope registration specification to express the corresponding relationship between the business scope entry and the industry classification in the catalog, and match the business scope entries corresponding to each industry classification According to the content of the activity, the industry form is obtained; combined with the corresponding relationship between the industry code and the standard items for starting a business, the industry form is associated with the standard items for starting a business, so as to realize the expansion of the industry form circle corresponding to each standard item for starting a business;

Step S32. Expansion of the event expression circle: Based on the historical registration data of individuals and enterprises, word segmentation is performed on the registered names of individuals and enterprises to obtain commonly used industry expressions, that is, colloquial expressions in enterprise registration names, combined with industry codes and business start-up Corresponding relationship of standard items, linking industry expressions with standard items for starting a business, so as to realize the expansion of the industry category circle corresponding to each standard item for starting a business;

Specifically, when segmenting individual or enterprise names in this step, use natural language processing technology to perform word segmentation processing on individual or enterprise registered names to obtain a list of keywords. For example, Hunan Kechuang Information Technology Co., Ltd., the word segmentation result is ["Hunan", "Kechuang", "Information Technology", "Co., Ltd"];

In this step, when extracting industry expressions, the most matching industry expressions are obtained by calculating the similarity between the keywords and the industry expressions in the industry expression dictionary table accumulated in the previous stage. The industry extracted from is expressed as "information technology";

Finally, this step combines the corresponding relationship between industry classification and business start-up standard items, and associates industry expressions with business start-up standard items, so as to realize the expansion of the industry category circle corresponding to each business start-up standard item;

Step S33. Expansion of the brand circle: Combining crawler and natural language processing technology, referring to the content of websites including China Brand Network, China Franchise Network, and Maigou.com to obtain brand data of various industries, and compare brand data with business start-up standards Correlation to form a complete multi-attribute knowledge base of business start-up matters.

Preferably, as shown in FIG. 5, step S33 specifically includes the following steps:

S331. Acquisition of brand data: obtain brand data from China Brand Network, China Franchise Network, and Maigou Network, and perform structured processing on the data;

S332. Matching of brand data and business start-up items: use natural language processing technology to realize automatic matching of brand data and business start-up items, and manually verify the matched data and associate unmatched data, thereby completing the brand to business start-up standard The mapping of items forms a complete multi-attribute knowledge base of business start-up items, and the above-mentioned knowledge base of business start-up items is synchronized to the search engine.

Preferably, as shown in Figure 6, the step S4 specifically includes the steps of:

S41. First, carry out full-text matching query of the input search keyword with the industry form, item expression and brand words in the multi-attribute knowledge base of business start-up matters. If they can be completely matched, then directly recommend the business start-up standard items associated with the keyword, such as, Entering "Tea Yanyue Se" will hit the brand circle attribute, and the business establishment item associated with this brand is "I want to open a beverage store", so the final recommendation result is "I want to open a beverage store";

S42. If no exact matching word is found, use natural language processing technology to perform word segmentation processing on the search keyword to obtain a search keyword group, use the error correction model to obtain the correct input search keyword, and then obtain the search keyword according to the synonym model A synonym group with similar words, for example, the input search term is "Guilin rice noodle", and the word segmentation result is ["Guilin", "rice noodle"], after the synonym model, no synonym is found for Guilin, but the synonym group obtained for rice noodle is ["rice noodle", "Rice Noodles", "Flour Noodles"...];

S43. Use the result of step S42 as a search keyword to search the standard items for starting a business and its multi-circle attributes, such as "Guilin rice noodles", word segmentation results ["Guilin", "rice noodles"], and synonym groups [ "Rice Noodles", "Rice Noodles", "Flour Noodles"...] are also used as search keywords to match the name of the standard item for starting a business, its industry form attribute, item expression attribute, brand attribute, etc. at the same time. The synonym "noodles" is matched to the standard item name "I want to open a noodle restaurant", and the synonym "rice noodles" is matched to the item expression attribute of the standard item "I want to open a breakfast shop"; according to the matching degree of the search keywords and the set search The weight order of keywords and attribute fields returns the highest-scoring N business start-up items as the recommended results of intelligent retrieval, such as the order of the final returned results is ["I want to open a noodle restaurant", "I want to open a breakfast restaurant", " I want to open a fast food restaurant", "I want to open a rice noodle processing factory"...].

Preferably, the step S43 specifically includes the steps of:

S431. When retrieving the catalog of standard matters for starting an enterprise, perform weight distribution, wherein, the weight Wt _i of the retrieval keyword is distributed as follows: the weight of the original retrieval keyword is Wt ₁ , and the word segmentation keyword obtained by the word segmentation algorithm in step S42 The weight of the phrase is Wt ₂ and the weight of the synonymous keyword phrase is Wt ₃ obtained according to the synonym model, and the weight of the retrieval keyword is set to Wt ₁ >Wt ₂ >Wt ₃ during retrieval; among them, for the search object enterprise to open a catalog of standard items and The distribution principle of the weight Ws _j of its multiple attributes is as follows: the weight of the standard item name obtained in step S2 is Ws ₁ , the weight of the industry form attribute obtained in step S31 is Ws ₂ , the weight of the item expression attribute obtained in step S32 is Ws ₃ , and step S33 The weight of the obtained brand attribute is Ws ₄ , and the weight of the retrieval target attribute is set to Ws ₁ >Ws ₂ >Ws ₃ >Ws ₄ during retrieval. For example, the weight of the synonym "noodles" as a retrieval keyword is Wt ₃ , the weight of matching the standard item name "I want to open a noodle restaurant" is WS ₁ ; the weight of the synonym "rice noodles" as a retrieval keyword is Wt ₃ , and The attribute of the standard item hit by the search keyword is an item expression attribute, so the attribute weight is Ws ₃ .

S432. The calculation algorithm for the score of the search term hitting the standard item is the traditional TF-IDF, and the score of the search term hitting the standard item is S _ij , wherein, i represents the source of the search term (i=1 indicates that the search term is the input search key word, i=2 represents that the search term is the participle result of the input search keyword, i=3 represents that the search term is a synonym obtained according to the synonym model); j represents the attribute circle layer that the search term hits (j=1 represents the The search term hits the standard item name, j=2 indicates that the search term hits the industry form attribute, j=3 indicates that the search term hits the item description attribute, and j=4 indicates that the search term hits the brand attribute) . For example, _S11 indicates the score of the input retrieval keyword hitting the name of the standard item; _S21 indicates the score of the word segmentation keyword hitting the name of the standard item; _S32 indicates the score of the synonymous keyword hitting the industry form attribute; _S34 indicates the synonymous key The score of the word hit brand attribute. For example, the standard item "I want to open a noodle restaurant" is obtained by matching the name of the item by searching the synonymous keyword "noodles", then the score of the search term "noodles" hitting the standard item is S ₃₁ .

Compute the final score for the criteria item being retrieved hits:

Among them, Wt _i is the weight of the search keywords, Ws _j is the weight of the catalog of standard items for starting a business and its multi-attributes, S _ij is the score of the search terms hitting the standard items, sorted and returned according to the final scores of each standard item retrieved The N items with the highest scores as the standard items for starting a business will be used as the recommended results of the intelligent search.

As shown in Figure 7, the preferred embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and operable on the processor, when the processor executes the program Steps for implementing the intelligent retrieval method for business start-up standard matters in the above-mentioned embodiment that integrates multiple attributes.

As shown in FIG. 8 , a preferred embodiment of the present application also provides a computer device, which may be a terminal or a living body detection server, and its internal structure may be shown in FIG. 8 . The computer device includes a processor, memory and a network interface connected by a system bus. Wherein, the processor of the computer device is used to provide calculation and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used to communicate with other external computer devices through a network connection. When the computer program is executed by the processor, the steps of the above-mentioned intelligent retrieval method for business start-up standard items integrated with multiple attributes can be realized.

Those skilled in the art can understand that the structure shown in FIG. 8 is only a block diagram of a partial structure related to the solution of this application, and does not constitute a limitation on the computer equipment to which the solution of this application is applied. The specific computer equipment can be More or fewer components than shown in the figures may be included, or some components may be combined, or have a different arrangement of components.

A preferred embodiment of the present application also provides a storage medium, the storage medium includes a stored program, and when the program is running, the device where the storage medium is located is controlled to execute the business start-up standard that integrates multiple attributes in the above embodiment The steps of the item intelligent retrieval method.

It should be noted that the steps shown in the flowcharts of the accompanying drawings may be performed in a computer system, such as a set of computer-executable instructions, and that although a logical order is shown in the flowcharts, in some cases, The steps shown or described may be performed in an order different than here.

If the functions described in the methods of this embodiment are implemented in the form of software function units and sold or used as independent products, they can be stored in one or more computing device-readable storage media. Based on this understanding, the part of the embodiment of the present application that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, the software product is stored in a storage medium, and includes several instructions to make a A computing device (which may be a personal computer, a server, a mobile computing device or a network device, etc.) executes all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), various media that can store program codes such as magnetic disk or optical disk. .

Those skilled in the art should understand that the embodiments of the present application may be provided as methods, systems, or computer program products. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein. The solutions in the embodiments of the present application can be realized by using various computer languages, for example, the object-oriented programming language Java and the literal translation scripting language JavaScript.

The present application is described with reference to flowcharts and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present application. It should be understood that each procedure and/or block in the flowchart and/or block diagram, and a combination of procedures and/or blocks in the flowchart and/or block diagram can be realized by computer program instructions. These computer program instructions may be provided to a general purpose computer, special purpose computer, embedded processor, or processor of other programmable data processing equipment to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing equipment produce a An apparatus for realizing the functions specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to operate in a specific manner, such that the instructions stored in the computer-readable memory produce an article of manufacture comprising instruction means, the instructions The device realizes the function specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions can also be loaded onto a computer or other programmable data processing device, causing a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process, thereby The instructions provide steps for implementing the functions specified in the flow chart or blocks of the flowchart and/or the block or blocks of the block diagrams.

While preferred embodiments of the present application have been described, additional changes and modifications to these embodiments can be made by those skilled in the art once the basic inventive concept is appreciated. Therefore, the appended claims are intended to be construed to cover the preferred embodiment and all changes and modifications which fall within the scope of the application.

Obviously, those skilled in the art can make various changes and modifications to the application without departing from the spirit and scope of the application. In this way, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application is also intended to include these modifications and variations.

Claims (9)

1. An enterprise development standard item intelligent retrieval method integrating multiple attributes is characterized by comprising the following steps:

s1, enterprise historical registration data preprocessing and statistical analysis: reading individual and enterprise historical registration data, cleaning and preprocessing the data, and then carrying out statistical analysis based on the processed data to obtain the occurrence frequency of each industry classification;

s2, generating an enterprise starting standard item catalog: based on national economy industry classification, generating an enterprise starting standard item catalog by combining individuals and enterprise historical registration data;

s3, constructing a multi-attribute circle layer of the enterprise starting standard items: defining the concept of a multi-attribute circle layer for the enterprise starting standard item, and acquiring enterprise starting related knowledge for each circle layer through a natural language processing technology to form a complete enterprise starting item multi-attribute knowledge base;

s4, outputting an intelligent retrieval result based on the enterprise starting item multi-attribute knowledge base: and acquiring a keyword list of the search keywords by means of a natural language processing technology and a synonym model, and determining matching degree scores of different items according to the attribute types hit by the keywords, so as to give intelligent search result recommendation of the enterprise starting items.

2. The method for intelligently retrieving the enterprise-sponsored standard event fused with the multivariate attributes as recited in claim 1, wherein the step S1 specifically comprises the steps of:

step S11, missing value processing: reading individual and enterprise historical registration data, and deleting data lacking a word name, namely lacking an enterprise name or an individual name in the data;

s12, grouping statistical analysis based on industry classification: and carrying out grouping statistics on the data based on the industry classification field in the original data to obtain the occurrence frequency of each industry classification.

3. The method for intelligently retrieving the enterprise-sponsored standard event fused with the multivariate attribute as claimed in claim 1, wherein the step S2 specifically comprises the steps of:

step S21, preliminarily forming an enterprise starting standard item catalog based on national standards: firstly, combining industry classification specific connotations and enterprise development related business knowledge based on industry fine classification in a national economy industry classification table to obtain an enterprise development standard item catalog preliminary version;

step S22, adjusting the enterprise to start a standard item catalog by combining regional characteristics: and (3) adjusting the preliminary version of the standard item catalog obtained in the step (S21) to reflect the time-space characteristics of the regional enterprise, and combining the regional individuals and the statistical analysis result of historical registration data of the enterprise within one year to form a complete standard item catalog of the enterprise.

4. The method for intelligently retrieving the enterprise-sponsored standard event fused with the multivariate attribute as claimed in claim 2, wherein the step S3 specifically comprises:

s31, expanding an industry form circle: firstly, the content of the operation range items is used for reference to expand the industry forms, namely, the corresponding relation between the operation range items and the industry classifications in the directory is expressed according to the operation range registration specifications, the operation range item activity content corresponding to each industry classification is matched, and the industry forms are obtained; then, the corresponding relation between the industry code and the enterprise starting standard item is combined, and the industry form is associated with the enterprise starting standard item, so that the expansion of the industry form circle corresponding to each enterprise starting standard item is realized;

step S32, expansion of item expression circle: based on historical registration data of individuals and enterprises, carrying out word segmentation processing on the individual and enterprise registration names to obtain common industry expressions, namely colloquial expressions in the enterprise registration names; associating the industry expression with the enterprise starting standard item by combining the corresponding relation between the industry code and the enterprise starting standard item, thereby realizing the expansion of the item expression circle corresponding to each enterprise starting standard item;

step S33, expansion of brand circles: combining crawler and natural language processing technology, obtaining brand data of each industry by using website contents including China brand network, china alliance network and purchase network for reference, and associating the brand data with enterprise starting standard items to form a complete enterprise starting item multi-attribute knowledge base.

5. The method for intelligently retrieving the enterprise-sponsored standard event fused with the multivariate attribute as claimed in claim 4, wherein the step S33 specifically comprises the steps of:

s331, obtaining brand data: acquiring brand data from a China brand network, a China alliance network and a purchase network, and carrying out structuring processing on the data;

s332, matching the brand data with the enterprise starting items: the automatic matching of brand data and enterprise starting items is realized by using a natural language processing technology, matched data is verified manually, and unmatched data is associated, so that the mapping from the brand to the enterprise starting standard items is completed, and a complete enterprise starting item multi-attribute knowledge base is formed.

6. The method for intelligently retrieving the enterprise-sponsored standard event fused with the multivariate attribute as claimed in claim 1, wherein the step S4 specifically comprises the steps of:

s41, firstly, carrying out full-text matching query on input search keywords and the operation forms, the item expressions and the brand words in the enterprise starting item multi-attribute knowledge base, and directly recommending enterprise starting standard items related to the keywords if complete matching is achieved;

s42, if the completely matched word cannot be found, performing word segmentation processing on the search keyword by using a natural language processing technology to obtain a search keyword group, obtaining a correct input search keyword by using an error correction model, and obtaining a synonym group similar to the search keyword according to the synonym model;

and S43, taking the result of the step S42 as a search keyword, searching the enterprise starting standard items and the multi-circle-layer attributes thereof, and returning N enterprise starting standard items with the highest scores as recommendation results of intelligent search according to the matching degree of the search keyword and the set weight sequence of the search keyword and the attribute field.

7. The method for intelligently retrieving the enterprise-sponsored standard event fused with the multivariate attribute as claimed in claim 6, wherein the step S43 specifically comprises the steps of:

s431, when searching the standard item catalog of the enterprise, distributing the weight, wherein the weight Wt of the search key word _i The distribution is as follows: the weight of the original search keyword is Wt ₁ And in step S42, the weight of the word group of the word segmentation keywords obtained by the word segmentation algorithm is Wt ₂ And obtaining the weight Wt of the synonymy keyword phrase according to the synonymy model ₃ Setting the weight of the search keyword to Wt during searching ₁ >Wt ₂ >Wt ₃ (ii) a Wherein, the standard item catalog and the weight Ws of the multi-component attribute are opened for the enterprise to be searched _j The distribution principle is as follows: the standard item name weight obtained in step S2 is Ws ₁ And the industry form attribute weight obtained in the step S31 is Ws ₂ The item expression attribute weight obtained in step S32 is Ws ₃ Step S33 obtains the weight of the brand attribute as Ws ₄ Setting the weight of the attribute of the retrieval target to Ws during retrieval ₁ >Ws ₂ >Ws ₃ >Ws ₄ ；

S432, the search word hits the standard itemThe score calculation algorithm is the traditional TF-IDF, and the score of the hit standard item of the search word is S _ij Wherein i represents a source of the search term, i =1 represents that the search term is an input search keyword, i =2 represents that the search term is a segmentation result of the input search keyword, and i =3 represents that the search term is a synonym obtained according to the synonym model; j represents the attribute circle layer of the search word hit, j =1 represents that the search word hit is a standard item name, j =2 represents that the search word hit is an industry form attribute, j =3 represents that the search word hit is an item expression attribute, j =4 represents that the search word hit is a brand attribute, wherein S is ₁₁ Then the score of the input search keyword hit standard item name is represented; s. the ₂₁ A score representing a hit of the participle keyword on the standard item name; s ₃₂ A score representing the hit of the synonymous keyword on the business form attribute; s. the ₃₄ A score representing a synonym keyword hit on a brand attribute;

calculating the final score of the retrieved hit of the standard item:

wherein, wt _i To retrieve the weight of a keyword, ws _j Weight, S, representing a catalog of enterprise-initiated standard events and its multiple attributes _ij And then representing the scores of the standard matters hit by the search words, and sorting according to the final scores of the standard matters obtained by searching to return the N enterprise starting standard matters with the highest scores as the recommendation result of the intelligent search.

8. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor,

the processor implements the steps of the intelligent searching method for the enterprise-initiated standard affairs fused with the multivariate attributes as recited in any one of claims 1 to 7 when executing the program.

9. A storage medium including a stored program, characterized in that,

controlling a device on which the storage medium is positioned to execute the steps of the enterprise-initiated standard event intelligent retrieval method fusing the multivariate attributes as claimed in any one of claims 1 to 7 when the program runs.

Images