Government affair item recommendation method and device, equipment, medium and program product

Abstract

The invention provides a government affair item recommendation method and device, equipment, a medium and a program product, and the method comprises the steps: carrying out the search based on a question inputted by a user, and obtaining candidate government affair items matched with the question inputted by the user; determining respective corresponding scores of the candidate government affair items in each dimension; based on the scores corresponding to the candidate government affair items in each dimension, comprehensive scores of the candidate government affair items are determined; determining a sorting result of the candidate government affair items based on the comprehensive scores of the candidate government affair items; and based on the sorting result of the candidate government affair items, determining a government affair item recommendation result corresponding to the question input by the user. The government affair item recommendation accuracy can be improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent government affairs technology, and in particular to a method, device, equipment, medium and program product for recommending government affairs matters. Background Art

[0002] With the rapid development of Internet technology, the digitization and onlineization of government information has continued to deepen, and the amount of information on various government affairs has increased dramatically. However, traditional search engines often rely on retrieval methods based on keyword matching, which cannot effectively meet users' efficient query needs in complex scenarios. Especially in the retrieval process of government affairs information, the information sources are wide and the content is complex. Traditional retrieval methods easily lead to disordered results and lack of pertinence, making it difficult for users to quickly find the most relevant content in the vast amount of information.

[0003] For example, when a user enters the query "I want to have a baby", traditional search methods may only match based on keywords, and the results often include relatively broad information such as "birth policy" and "birth insurance", but do not accurately match the specific item of "birth service registration card" that the user actually needs. This single keyword matching method is not only prone to generate a large amount of irrelevant information, but may also greatly reduce the efficiency of user information retrieval. Therefore, the accuracy of government affairs recommendations is currently low. Summary of the invention

[0004] The present application provides a government affairs matter recommendation method, device, equipment, medium and program product for improving the accuracy of government affairs matter recommendation.

[0005] According to a first aspect of an embodiment of the present application, a method for recommending government affairs is provided, comprising:

[0006] Searching based on the question input by the user to obtain candidate government affairs items that match the question input by the user;

[0007] Determine the scores corresponding to the candidate government affairs in each dimension;

[0008] Determining a comprehensive score of the candidate government affairs item based on the scores corresponding to each dimension of the candidate government affairs item;

[0009] Determining a ranking result of the candidate government affairs matters based on the comprehensive scores of the candidate government affairs matters;

[0010] Based on the ranking results of the candidate government affairs, a government affairs matter recommendation result corresponding to the question input by the user is determined.

[0011] Optionally, the dimension includes a data source dimension;

[0012] Determining the scores corresponding to the candidate government affairs in various dimensions includes:

[0013] Determine a basic score of the data source of the candidate government affairs matter; wherein the basic score is positively correlated with the degree of correlation between the data source and the handling of the government affairs matter;

[0014] Determining a credibility score of the data source of the candidate government affairs matter;

[0015] Based on the basic score and the credibility score, the score of the candidate government affairs item in the data source dimension is determined.

[0016] Optionally, the dimension includes a relevance dimension; wherein the relevance dimension is used to characterize the relevance between the candidate government affairs item and the user input question;

[0017] Determining the scores corresponding to the candidate government affairs in various dimensions includes:

[0018] Extracting each keyword in the question input by the user;

[0019] For any keyword in the question input by the user, perform the following operations:

[0020] Determining a first text match between the any one keyword and the name of the candidate government affairs item;

[0021] Determining a second text matching degree between the any one keyword and the specific content of the candidate government affairs item;

[0022] Determining a matching degree between the any one keyword and the candidate government affairs item based on the first text matching degree and the second text matching degree;

[0023] Based on the matching degree between each keyword in the question input by the user and the candidate government affairs, the score of the candidate government affairs in the relevance dimension is determined.

[0024] Optionally, the dimension includes a hit event label dimension;

[0025] Determining the scores corresponding to the candidate government affairs in various dimensions includes:

[0026] Determine each target tag among the tags of the candidate government affairs matters that matches the question input by the user;

[0027] Based on each of the target tags, the score of the candidate government affairs item in the hit item tag dimension is determined.

[0028] Optionally, the dimension includes a matter popularity dimension;

[0029] Determining the scores corresponding to the candidate government affairs in various dimensions includes:

[0030] At first preset time intervals, obtaining the query popularity of the candidate government affairs items;

[0031] Determining a popularity score of the candidate government affairs matter based on the query popularity;

[0032] At every second preset time, the heat score is decayed according to a preset decay ratio to obtain a heat score after decay;

[0033] Based on the attenuated heat score, the score of the candidate government affairs item in the item heat dimension is determined.

[0034] Optionally, the dimension includes a release time dimension;

[0035] Determining the scores corresponding to the candidate government affairs in various dimensions includes:

[0036] Determining the release interval of the candidate government affairs item based on the release time of the candidate government affairs item and the current time;

[0037] Based on the release interval of the candidate government affairs, the score of the candidate government affairs in the release time dimension is determined.

[0038] Optionally, the dimension includes a matter evaluation dimension;

[0039] Determining the scores corresponding to the candidate government affairs in various dimensions includes:

[0040] Based on the user service evaluation of the candidate government affairs item, the score of the candidate government affairs item in the item evaluation dimension is determined.

[0041] Optionally, the dimension includes a handling frequency dimension;

[0042] Determining the scores corresponding to the candidate government affairs in various dimensions includes:

[0043] Based on the handling frequency of the candidate government affairs items within a preset time period, the score of the candidate government affairs items in the handling frequency dimension is determined.

[0044] Optionally, the dimensions include data source dimension, relevance dimension, hit item label dimension, item popularity dimension, release time dimension, item evaluation dimension and handling frequency dimension; wherein the relevance dimension is used to characterize the relevance between the candidate government affairs item and the user input question;

[0045] The corresponding scores in each dimension include the score in the data source dimension, the score in the relevance dimension, the score in the hit matter label dimension, the score in the matter popularity dimension, the score in the release time dimension, the score in the matter evaluation dimension, and the score in the handling frequency dimension.

[0046] Optionally, determining the comprehensive score of the candidate government affairs item based on the scores of the candidate government affairs item in each dimension includes:

[0047] Determine the weights corresponding to each dimension;

[0048] Based on the scores corresponding to the candidate government affairs in each dimension and the weights corresponding to each dimension, a weighted sum is performed to obtain a comprehensive score of the candidate government affairs.

[0049] According to a second aspect of an embodiment of the present application, a government affairs recommendation device is provided, including:

[0050] A search unit, configured to search based on a question input by a user, and obtain candidate government affairs items that match the question input by the user;

[0051] A first processing unit is used to determine the scores corresponding to the candidate government affairs in each dimension;

[0052] A second processing unit, configured to determine a comprehensive score of the candidate government affairs item based on the scores corresponding to the candidate government affairs items in each dimension;

[0053] A ranking unit, configured to determine a ranking result of the candidate government affairs items based on the comprehensive scores of the candidate government affairs items;

[0054] The recommendation unit is used to determine the government affairs item recommendation result corresponding to the user input question based on the ranking result of the candidate government affairs items.

[0055] According to a third aspect of an embodiment of the present application, there is provided an electronic device, including a memory and a processor;

[0056] The memory is connected to the processor and is used to store programs;

[0057] The processor is used to implement the government affairs recommendation method as described in the first aspect by running the program in the memory.

[0058] According to a fourth aspect of an embodiment of the present application, a storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method for recommending government affairs matters as described in the first aspect is implemented.

[0059] According to a fifth aspect of an embodiment of the present application, a computer program product is provided, comprising computer program instructions, which, when executed by a processor, enable the processor to execute the government affairs recommendation method as described in the first aspect.

[0060] In this application, a search is performed based on a user input question to obtain candidate government affairs items that match the user input question, and the scores corresponding to the candidate government affairs items in each dimension are determined. The candidate government affairs items that match the user input question are not directly recommended to the user, but the scores corresponding to the candidate government affairs items in each dimension are determined, and the candidate government affairs items are scored through multiple dimensions. Then, based on the scores corresponding to the candidate government affairs items in each dimension, the comprehensive scores of the candidate government affairs items are determined. Compared with the traditional single keyword search, the comprehensive scores of the candidate government affairs items can better reflect the degree of match between the candidate government affairs items and the actual needs of the user. Based on the comprehensive scores of the candidate government affairs items, the ranking results of the candidate government affairs items are determined, and the candidate government affairs items are ranked based on the comprehensive scores of the candidate government affairs items. Then, based on the ranking results of the candidate government affairs items, the recommended results of the government affairs items corresponding to the user input question are determined. On the basis that the comprehensive scores of the candidate government affairs items can better reflect the degree of match between the candidate government affairs items and the actual needs of the user, the government affairs items recommendation results can be more accurately matched to the actual needs of the user by recommending after sorting, further improving the accuracy of government affairs item recommendations, and improving the efficiency and satisfaction of users in finding government affairs items. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0062] Figure 1 A flowchart of a method for recommending government affairs provided in an embodiment of the present application;

[0063] Figure 2 A schematic diagram of a process for determining the score of a candidate government affairs item in a data source dimension provided in an embodiment of the present application;

[0064] Figure 3 A schematic diagram of a process for determining the score of a candidate government affairs item in a relevance dimension provided in an embodiment of the present application;

[0065] Figure 4A schematic diagram of a process for determining the score of a candidate government affairs item in the hit item label dimension provided in an embodiment of the present application;

[0066] Figure 5 A schematic diagram of a process for determining the score of a candidate government affairs matter in the matter heat dimension provided in an embodiment of the present application;

[0067] Figure 6 A schematic diagram of a process for determining the score of a candidate government affairs item in the dimension of release time provided in an embodiment of the present application;

[0068] Figure 7 A flowchart of a method for recommending government affairs provided in an embodiment of the present application;

[0069] Figure 8 A schematic diagram of a process of step 103 provided in an embodiment of the present application;

[0070] Fig. 9 A schematic diagram of the structure of a government affairs recommendation device provided in an embodiment of the present application;

[0071] Fig.10 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0072] At present, in the process of retrieving government affairs information, traditional search engines often rely on retrieval methods based on keyword matching. Such methods are usually based on text matching between keywords in user queries and government affairs content, and determine similarity by calculating the frequency or position of keywords in the text. Commonly used retrieval methods based on keyword matching are as follows:

[0073] Boolean model: Both the query and the document are represented as a collection of keywords, and then Boolean logic (such as AND, OR, NOT, etc.) is used to determine whether the document matches the query. This method is simple and intuitive, but it usually does not take into account the frequency or importance of keywords.

[0074] Multi-keyword sorting: When multiple keywords need to be sorted, the composite keyword sorting method is used. This method combines multiple keywords into a composite keyword and sorts them according to the size of the composite keyword.

[0075] Vector Space Model: Each document and query is represented as a vector in a high-dimensional space, where each dimension corresponds to a keyword. By calculating the similarity between the document vector and the query vector (such as cosine similarity), the relevance of the document to the query can be determined.

[0076] Term Frequency-Inverse Document Frequency (TF-IDF) algorithm: This is a statistical method used to evaluate the importance of a term to a document set or a document in a corpus. The importance of a term increases in direct proportion to the number of times it appears in a document, but decreases in inverse proportion to the frequency of its appearance in the corpus.

[0077] Existing government information retrieval solutions usually use keyword matching technology to perform a simple lexical comparison between user queries and government affairs content, and lack the ability to understand semantics. The disadvantage of this method is that it cannot accurately capture the user's query intent, which can easily lead to generalization or irrelevant search results. For example, when a user searches for "I want to have a baby", the system may return general results containing keywords such as "birth policy" or "birth insurance", but cannot accurately match the specific demand "birth service registration card". The existing technology lacks in-depth analysis of query semantics, resulting in insufficient accuracy of search results. Lack of semantic understanding and precise matching capabilities.

[0078] Therefore, the accuracy of government affairs recommendations is currently low.

[0079] To solve this problem, existing technologies have begun to introduce more context understanding, semantic matching and user intent recognition methods, but these methods still face challenges such as high computational complexity, low accuracy and insufficient information matching.

[0080] In order to improve the accuracy of government affairs recommendation, the present application provides a government affairs recommendation method, device, equipment, medium and program product.

[0081] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0082] Exemplary Implementation Environment

[0083] The government affairs recommendation method according to the embodiment of the present application can be executed by an electronic device such as a terminal device or a server. The terminal device can be a user device, a mobile device, a computing device, a wearable device, etc. The server can be an independent physical server, a server cluster composed of multiple physical servers, or a cloud server capable of cloud computing. The method can be implemented by a processor calling a computer-readable program instruction stored in a memory.

[0084] Exemplary Methods

[0085] Please refer to Figure 1 , in an exemplary embodiment, a government affair item recommendation method is provided. As Figure 1 shown, the process of the government affair item recommendation method mainly includes:

[0086] Step 101, search based on the user input question to obtain candidate government affair items that match the user input question.

[0087] In an exemplary embodiment, the user enters the user input question in a search engine. For example, the user input question is "I want to have a baby".

[0088] In an exemplary embodiment, step 101 includes: preprocessing the user input question to obtain each keyword in the user input question; based on each keyword in the user input question and the document to be retrieved, obtain candidate government affair items that match the user input question from the document to be retrieved.

[0089] In an exemplary embodiment, preprocessing the user input question to obtain each keyword in the user input question may include: performing word segmentation and stop word removal on the user input question to obtain each keyword in the user input question. Word segmentation refers to splitting the input string of the user (i.e., the user input question) into individual words. For example: if the user input question is "I want to have a baby", after word segmentation, it is ["I", "have a baby"]. Stop word removal refers to removing common but meaningless words (such as "of", "and", etc.).

[0090] In an exemplary embodiment, based on each keyword in the user input question and the document to be retrieved, obtaining candidate government affair items that match the user input question from the document to be retrieved may include: for any one keyword in the user input question, obtain the target document containing any one keyword in the user input question from the document to be retrieved; through Boolean model matching and / or vector space model matching, obtain candidate government affair items that match the user input question from each target document.

[0091] In an exemplary embodiment, documents to be retrieved related to the user input question are collected from a database, the content of the documents to be retrieved is parsed and text information is extracted, a target document containing any keyword in the user input question is obtained from the documents to be retrieved, a document list containing the word is established for any keyword in the user input question, and the user input question is converted into a space vector to obtain a query vector. For example, the user input question is "I want to have a baby", which is converted into a space vector, and the query vector obtained is [1,1]. Boolean model matching: matching documents according to Boolean logic (AND, OR, NOT). For example: query "I AND have a baby", matching documents containing "I" and "have a baby". Vector space model matching: calculating the similarity between the query vector and the document vector. Candidate government affairs items that match the user input question can be obtained only by using Boolean model matching; can also be obtained by using vector space model matching only; can also be obtained by using other matching methods to obtain candidate government affairs items that match the user input question; can also be obtained by using Boolean model matching and vector space model matching to obtain candidate government affairs items that match the user input question; this application is not limited to this.

[0092] In an exemplary embodiment, the candidate government affairs items that match the user input question are unranked intermediate search results.

[0093] Step 102, determining the scores corresponding to the candidate government affairs in each dimension.

[0094] In some embodiments, the scores corresponding to the candidate government affairs items in each dimension are determined by performing rapid scoring in a parallel computing manner to obtain the scores corresponding to the candidate government affairs items in each dimension.

[0095] For example, parallel computing may refer to the parallel processing of the calculation of scores of a candidate government matter in different dimensions. The parallel processing of the calculation of scores of different candidate government matters can improve the calculation efficiency of the scores corresponding to the candidate government matters in each dimension.

[0096] In an exemplary embodiment, candidate government affairs may be obtained from multiple government affairs-related data sources, including but not limited to government services, public services, policy documents, business lists and other data sources, to ensure that all types of government affairs that users may need are covered.

[0097] In some embodiments, the dimension includes a data source dimension.

[0098] In some embodiments, Figure 2 As shown in the figure, the scores of candidate government affairs in the data source dimension are determined, including:

[0099] Step 201, determine the basic score of the data source of the candidate government affairs.

[0100] Among them, the basic score is positively correlated with the correlation between the data source and the handling of government affairs.

[0101] In the exemplary embodiment, the data sources of candidate government affairs items are assigned different basic scores according to the type of data sources, reflecting the weights of different data sources. The data sources of candidate government affairs items are special. For example, data from the "implementation list" has a higher priority because it is directly related to the business handling of government affairs items, and is assigned 10 points; data from the "convenience services" source provides convenient matters commonly used by users, and is assigned 8 points; data from the "service interaction" source contains interactive data between users and the government, and is assigned 6 points; data from the "policies and regulations" source helps users understand the policy background, and is assigned 5 points.

[0102] The basic score is positively correlated with the correlation between the data source and the handling of government affairs. The higher the correlation between the data source and the handling of government affairs, the higher the basic score, and thus the higher the priority of the data source. Through this classification, it is ensured that candidate government affairs from high-priority data sources occupy a higher position in the ranking.

[0103] Step 202, determine the credibility score of the data source of the candidate government affairs matter.

[0104] In an exemplary embodiment, for data sources certified by authoritative institutions (such as government departments at all levels and certified government platforms), 5 points are added to reflect the credibility and authority of the data. That is, the credibility score of data sources certified by authoritative institutions is 5 points, and the credibility score of data sources not certified by authoritative institutions is 0 points.

[0105] The step of determining the credibility score of the data source of candidate government affairs is particularly important in the ranking of government affairs. It helps to improve the reliability of the ranking results and allows users to see high-quality authoritative information first.

[0106] Step 203, based on the basic score and the credibility score, determine the score of the candidate government affairs item in the data source dimension.

[0107] In an exemplary embodiment, step 203 may include directly adding the basic score and the credibility score to obtain the score of the candidate government affairs item in the data source dimension; step 203 may also include weighted summing the basic score and the credibility score to obtain the score of the candidate government affairs item in the data source dimension; step 203 may also have other implementation methods, and this application is not limited to this.

[0108] In an exemplary embodiment, similar candidate government affairs items from different data sources are identified and deduplicated, thereby improving the simplicity and practicality of the presentation of recommendation results and enabling efficient integration of multi-source data.

[0109] At present, government information comes from various sources, but most of the existing sorting methods do not fully consider the weight and credibility of different data sources, and only perform basic sorting based on data categories or information types, resulting in low-quality or redundant content in search results. The reliability of government information has a huge impact on user experience, and the lack of weighting for authoritative institutions or high-quality sources affects users' trust in search results.

[0110] In this application, the basic score of the data source of the candidate government affairs is determined, where the basic score is positively correlated with the degree of correlation between the data source and the handling of government affairs, and the credibility score of the data source of the candidate government affairs is determined. Based on the basic score and the credibility score, the score of the candidate government affairs in the data source dimension is determined. By allocating weights to different data sources and using a credibility scoring mechanism, it is ensured that the content of highly authoritative and high-quality data sources is prioritized, thereby providing users with more reliable information, so that authoritatively certified high-quality data is prioritized in the sorting, avoiding interference from low-quality or duplicate information, enhancing users' trust in search results, and ensuring that high-quality information from authoritative data sources is prioritized.

[0111] In some embodiments, the dimension includes a relevance dimension; wherein the relevance dimension is used to characterize the relevance between the candidate government affairs matters and the user input question.

[0112] In some embodiments, Figure 3 As shown in the figure, determine the scores of candidate government affairs in the relevance dimension, including:

[0113] Step 301: extract keywords from the question input by the user.

[0114] Step 302: for any keyword in the question input by the user, perform the following operations: determine the first text matching degree between any keyword and the name of the candidate government affairs item.

[0115] Step 303: determine a second text matching degree between any keyword and the specific content of the candidate government affairs item.

[0116] Step 304: Determine the matching degree between any keyword and the candidate government affairs item based on the first text matching degree and the second text matching degree.

[0117] In an exemplary embodiment, a first weight may be set for the first text matching degree, and a second weight may be set for the second text matching degree. Based on the first text matching degree, the first weight, the second text matching degree, and the second weight, a weighted sum is performed to obtain the matching degree between any keyword and the candidate government affairs item. For example, the name of the candidate government affairs item has a higher weight because it directly expresses the user's needs, and the first weight may be set to 0.7; the specific content of the candidate government affairs item provides a detailed description of the candidate government affairs item, and the second weight may be set to 0.3. This design ensures that candidate government affairs items with higher name matching degrees are displayed first.

[0118] Step 305: Determine the score of the candidate government affairs item in the relevance dimension based on the matching degree between each keyword in the question input by the user and the candidate government affairs item.

[0119] In some embodiments, when the user input question contains only one keyword, the matching degree between the keyword and the candidate government affairs item is directly determined as the score of the candidate government affairs item in the relevance dimension.

[0120] In other embodiments, when the user input question contains multiple keywords, each keyword corresponds to a matching degree, and the matching degrees between each keyword in the user input question and the candidate government affairs items are averaged to obtain an average matching degree, and the average matching degree is determined as the score of the candidate government affairs item in the relevance dimension. This method ensures that for multi-keyword queries, each keyword can be comprehensively matched, thereby improving the overall relevance of the query results.

[0121] Extract each keyword in the user input question, and for any keyword in the user input question, perform the following operations: determine the first text matching degree between any keyword and the name of the candidate government affairs item, determine the second text matching degree between any keyword and the specific content of the candidate government affairs item, determine the matching degree between any keyword and the candidate government affairs item based on the first text matching degree and the second text matching degree, and determine the score of the candidate government affairs item on the relevance dimension based on the matching degree between each keyword in the user input question and the candidate government affairs item. By calculating the first text matching degree between each keyword in the user input question and the name of the candidate government affairs item, and the second text matching degree between each keyword and the specific content of the candidate government affairs item, the semantic matching accuracy of the search results is improved, the user's query intent can be accurately identified, the potential intent of the user's query can be understood, and the accuracy of the search results is improved. For example, when the user enters the query "I want to have a baby", the system can intelligently identify and give priority to recommending precise items such as "birth service registration card", achieving accurate matching of semantics and intent, and then achieving accurate recommendation.

[0122] In some embodiments, the dimension includes a hit label dimension.

[0123] In some embodiments, Figure 4 As shown, the scores of candidate government affairs items on the hit item label dimension are determined, including:

[0124] Step 401, determining each target tag among the tags of the candidate government affairs items that matches the question input by the user.

[0125] In an exemplary embodiment, the labels of candidate government affairs include primary labels and secondary labels. The primary labels of candidate government affairs usually include the core business areas of the candidate government affairs, helping the system to prioritize the display of primary matters that meet user needs. Secondary labels enable the system to identify relevant areas of user queries and provide relevant content as recommendations. Primary labels can be key labels for the business areas of candidate government affairs, and secondary labels can be auxiliary labels for the relevant business areas of candidate government affairs. For example, the candidate government affairs item is "Family Service Registration Card", the primary label is "Birth Registration", and the secondary labels are "Social Security Application", "Hukou Registration", etc. Primary labels are labels that are directly related to the candidate government affairs items, and secondary labels are labels that have a certain correlation with the candidate government affairs items, but are not directly related.

[0126] Step 402, based on each target tag, determine the score of the candidate government affairs item in the hit item tag dimension.

[0127] For example, a major tag in the target tag adds 20 points, and a minor tag in the target tag adds 10 points. The scores of candidate government affairs items in the hit item tag dimension are obtained by weighted accumulation according to the importance and number of tags. This allows candidate government affairs items with more relevant tags to have higher priority in the sorting.

[0128] Determine the target tags in the candidate government affairs tags that match the user input question, and determine the scores of the candidate government affairs in the hit item tag dimension based on the target tags. Through the tag hit mechanism, the system can understand the potential intent of the user's query and improve the accuracy of the search results. For example, when the user enters the query "I want to have a baby", the system can intelligently identify and prioritize precise items such as "Fertility Service Registration Card" to achieve accurate matching of semantics and intent.

[0129] In some embodiments, the dimension includes a matter popularity dimension.

[0130] In some embodiments, determining the score of a candidate government matter on the matter heat dimension includes: obtaining the query heat of the candidate government matter at first preset time intervals; determining the heat score of the candidate government matter based on the query heat; and determining the score of the candidate government matter on the matter heat dimension based on the heat score of the candidate government matter.

[0131] In an exemplary embodiment, the popularity score of a candidate government matter is determined based on the query popularity. The candidate government matters may be ranked based on their query popularity to obtain a query popularity ranking of the candidate government matters; and the popularity score of the candidate government matter is determined based on the query popularity ranking of the candidate government matters.

[0132] For example, different scores are assigned based on the query popularity ranking in the past month: 50 points are added to the top 30 candidate government affairs items, 30 points are added to the candidate government affairs items ranked 31-99, and 10 points are added to the candidate government affairs items ranked 100-199. This ensures that users can quickly obtain the current popular items.

[0133] In some embodiments, Figure 5 As shown in the figure, the scores of candidate government affairs items in terms of item popularity are determined, including:

[0134] Step 501, obtaining the query popularity of the candidate government affairs items at first preset time intervals.

[0135] At first preset time intervals, the query popularity of candidate government affairs items is obtained, and the query popularity of candidate government affairs items is regularly counted to ensure that users can quickly obtain current popular items.

[0136] Step 502, based on the query popularity, determine the popularity score of the candidate government affairs matter.

[0137] In an exemplary embodiment, the popularity score of a candidate government matter is determined based on the query popularity. The candidate government matters may be ranked based on their query popularity to obtain a query popularity ranking of the candidate government matters; and the popularity score of the candidate government matter is determined based on the query popularity ranking of the candidate government matters.

[0138] For example, different scores are assigned based on the query popularity ranking in the past month: 50 points are added to the top 30 candidate government affairs items, 30 points are added to the candidate government affairs items ranked 31-99, and 10 points are added to the candidate government affairs items ranked 100-199. This ensures that users can quickly obtain the current popular items.

[0139] Step 503: attenuate the heat score according to a preset attenuation ratio every second preset time period to obtain a heat score after attenuation.

[0140] In an exemplary embodiment, the preset decay ratio can be set to 10% or other values, and this application is not limited to this. To prevent outdated candidate government affairs from occupying high priority positions, the heat score decays by 10% every second preset time to ensure that the latest needs are given priority and dynamically reflect the changing trends of user concerns.

[0141] Step 504, based on the attenuated heat score, determine the score of the candidate government affairs item in the item heat dimension.

[0142] At present, in the search for government information, timeliness and popularity of matters are often the core factors that users pay attention to. Existing technologies usually do not introduce dynamic factors such as search popularity, and lack a time decay mechanism. This results in some outdated content still occupying an important position in the ranking, while matters with high recent search popularity may be ignored, which cannot meet users' needs for timely and high-frequency information. The lack of dynamic tracking of popularity also makes it difficult for the system to adapt to the ever-changing needs of users.

[0143] In the present application, the query heat of the candidate government affairs items is obtained every first preset time period, and the heat score of the candidate government affairs items is determined based on the query heat. Every second preset time period, the heat score is decayed according to a preset decay ratio to obtain the attenuated heat score, and the score of the candidate government affairs items in the item heat dimension is determined based on the attenuated heat score. By introducing query heat and time decay, the sorting weights of recent high-heat items in the search results can be dynamically adjusted to give priority to displaying recent high-heat or newly released item information, helping users quickly locate content that is currently highly concerned and highly timely. At the same time, the system will take time decay into account to avoid outdated information occupying a prominent position, so that the sorting results always match the user's latest needs. It ensures that users can quickly find items that are currently of high concern, while avoiding interference from outdated content, and optimizing the timeliness of the results.

[0144] In some embodiments, the dimensions include a publishing time dimension.

[0145] In some embodiments, Figure 6 As shown in the figure, the score of the candidate government affairs item in the release time dimension is determined, including:

[0146] Step 601, based on the release time of the candidate government affairs item and the current time, determine the release interval of the candidate government affairs item.

[0147] In an exemplary embodiment, the difference obtained by subtracting the release time of the candidate government affairs item from the current time is determined as the release interval duration of the candidate government affairs item.

[0148] Step 602: Determine the score of the candidate government affairs item in the dimension of release time based on the release interval of the candidate government affairs item.

[0149] For example, if the interval between the release of candidate government affairs is within 6 months, 20 points will be added to support users to find the latest and most timely information. If the interval between the release of candidate government affairs is between 6 and 12 months, 10 points will be added to reflect its certain reference value. If the interval between the release of candidate government affairs is more than 12 months, no points will be added to maintain the timeliness of the information.

[0150] Based on the release time and current time of the candidate government affairs items, the release interval of the candidate government affairs items is determined. Based on the release interval of the candidate government affairs items, the score of the candidate government affairs items in the release time dimension is determined to support users in finding the latest and most timely information.

[0151] In some embodiments, the dimension includes an issue evaluation dimension. Determining the score of the candidate government affairs issue on the issue evaluation dimension includes: determining the score of the candidate government affairs issue on the issue evaluation dimension based on the user service evaluation of the candidate government affairs issue.

[0152] For example, if the user service rating of a candidate government matter is 4 stars or above, 20 points will be added and matters with good user experience will be displayed first; if the user service rating of a candidate government matter is between 3-4 stars, 10 points will be added and matters with medium ratings will be displayed appropriately; if the user service rating of a candidate government matter is lower than 3 stars, no points will be added.

[0153] Currently, user feedback is an important indicator for improving sorting quality, but existing sorting schemes lack the integration of user evaluation and service feedback. User satisfaction with items often reflects the quality and practicality of items. Failure to use the mechanism of giving priority to high-rated items may result in low-rated items being mixed with high-quality items, affecting user experience and service efficiency.

[0154] In this application, the scores of candidate government affairs items in the item evaluation dimension are determined based on the user service evaluation of the candidate government affairs items. User evaluation is an important indicator for judging the service quality of items. User evaluation scores are added to the sorting process, so that high-quality items with good user evaluations can get higher rankings, thereby effectively integrating user feedback into the sorting mechanism. This mechanism not only helps to prioritize the display of high-quality information, but also helps users quickly filter out items with high recognition from other users. Using user evaluation as one of the key indicators for sorting, highly rated and highly satisfied items are given priority, so that the system can adjust the sorting with the help of real feedback from users. By introducing an evaluation mechanism, users can find items with higher experience and satisfaction more quickly, which improves the practicality of the system and user experience.

[0155] In some embodiments, the dimension includes a handling frequency dimension. Determining the score of the candidate government affairs item on the handling frequency dimension includes: determining the score of the candidate government affairs item on the handling frequency dimension based on the handling frequency of the candidate government affairs item within a preset time period.

[0156] For example, if a candidate government item is one of the top 10 most frequently handled items in the past three months, 30 points will be added. High-frequency items reflect user needs, and the system gives priority to display, simplifying the process for users to find high-frequency items. If a candidate government item is one of the top 11-50 most frequently handled items in the past three months, 20 points will be added. If a candidate government item is one of the top 51-100 most frequently handled items in the past three months, 10 points will be added to ensure that frequently used items can be appropriately prioritized.

[0157] At present, some government affairs have become high-frequency affairs due to the daily needs of the public. Most existing sorting technologies do not take high-frequency affairs into consideration, resulting in the failure to prioritize the affairs that users need to query repeatedly in the search results, which increases the query time and operation steps of users and lacks the ability to understand the frequency of user needs.

[0158] In this application, the scores of candidate government affairs items in the dimension of frequency of handling are determined based on the frequency of handling of candidate government affairs items within a preset time period. For matters that are frequently handled by the public, analysis is conducted based on recent case handling data to identify the most frequently queried and used matters in the public's service needs, and high-frequency matters are included in the content with higher sorting priority, helping users to quickly find matters that other users frequently pay attention to and handle, optimizing the practicality and service orientation of information, and providing users with recommendations for matters that are more in line with daily needs. Greatly shorten the time cost of users looking for high-frequency matters and improve query efficiency.

[0159] In some embodiments, the dimensions include data source dimension, relevance dimension, hit item label dimension, item popularity dimension, release time dimension, item evaluation dimension, and handling frequency dimension; wherein the relevance dimension is used to characterize the relevance between the candidate government affairs items and the user input question;

[0160] The corresponding scores in each dimension include the score in the data source dimension, the score in the relevance dimension, the score in the hit matter label dimension, the score in the matter popularity dimension, the score in the release time dimension, the score in the matter evaluation dimension, and the score in the handling frequency dimension.

[0161] In some embodiments, Figure 7 As shown in the figure, the process of the government affairs recommendation method mainly includes:

[0162] Step 701, search based on the question input by the user to obtain candidate government affairs items that match the question input by the user.

[0163] Step 702, determine the score of the candidate government affairs item in the data source dimension, the score in the relevance dimension, the score in the hit item label dimension, the score in the item popularity dimension, the score in the release time dimension, the score in the item evaluation dimension, and the score in the handling frequency dimension.

[0164] In an exemplary embodiment, the scores of candidate government affairs in the data source dimension can be determined by referring to Figure 2 , to determine the scores of candidate government affairs in the relevance dimension, please refer to Figure 3 , to determine the scores of candidate government affairs in the hit matter label dimension, please refer to Figure 4 To determine the scores of candidate government affairs in terms of the issue popularity dimension, please refer to Figure 5 To determine the scores of candidate government affairs in terms of release time, please refer to Figure 6 Determining the score of the candidate government affairs item on the item evaluation dimension includes: determining the score of the candidate government affairs item on the item evaluation dimension based on the user service evaluation of the candidate government affairs item. Determining the score of the candidate government affairs item on the handling frequency dimension includes: determining the score of the candidate government affairs item on the handling frequency dimension based on the handling frequency of the candidate government affairs item within a preset time period.

[0165] Step 703, based on the candidate government affairs' score in the data source dimension, the score in the relevance dimension, the score in the hit affair tag dimension, the score in the affair popularity dimension, the score in the release time dimension, the score in the affair evaluation dimension, and the score in the handling frequency dimension, determine the comprehensive score of the candidate government affairs.

[0166] Step 704, determining the ranking results of the candidate government affairs items based on the comprehensive scores of the candidate government affairs items.

[0167] Step 705: Based on the ranking results of the candidate government affairs, determine the government affairs recommendation results corresponding to the question input by the user.

[0168] At present, most of the sorting methods in the existing technology are based on a single dimension (such as keywords, time or evaluation) and lack a multi-dimensional comprehensive sorting mechanism. In complex government information retrieval scenarios, it is difficult to effectively meet the diverse needs of users by relying solely on a single dimension, especially when faced with a large amount of heterogeneous data. The sorting results often fail to accurately display the most relevant and high-quality information, reducing the user experience.

[0169] In this application, the candidate government affairs items are scored through multiple dimensions, and then the comprehensive scores of the candidate government affairs items are determined based on the scores corresponding to the candidate government affairs items in each dimension. Compared with the traditional single keyword search, the comprehensive scores of the candidate government affairs items can better reflect the degree of match between the candidate government affairs items and the actual needs of users. Based on the comprehensive scores of the candidate government affairs items, the ranking results of the candidate government affairs items are determined, and the candidate government affairs items are ranked based on the comprehensive scores of the candidate government affairs items. Then, based on the ranking results of the candidate government affairs items, the government affairs item recommendation results corresponding to the user input questions are determined. On the basis that the comprehensive scores of the candidate government affairs items can better reflect the degree of match between the candidate government affairs items and the actual needs of users, the government affairs item recommendation results can be more accurately matched to the actual needs of users by recommending after sorting, further improving the accuracy of government affairs item recommendations, and improving the efficiency and satisfaction of users in finding government affairs items.

[0170] By comprehensively considering multiple dimensions to weight the search results, it is not only based on keyword matching, but also can effectively identify users' potential needs and query intentions. Compared with traditional single keyword retrieval, it can more accurately match users' actual needs and improve the accuracy of search results. For example, when a user searches for "I want to have a baby", the system can intelligently identify and prioritize specific items such as "Fertility Service Registration Card" that meet user needs. Taking into account multiple dimensions such as data source dimension, relevance dimension, hit item label dimension, item popularity dimension, release time dimension, item evaluation dimension and handling frequency dimension, an intelligent sorting solution suitable for complex scenarios is formed. It not only improves the applicability of the system, but also greatly enhances the personalization of search results to adapt to the diverse demand scenarios of government affairs.

[0171] Step 103: Determine the comprehensive score of the candidate government affairs item based on the scores of the candidate government affairs item in each dimension.

[0172] In some embodiments, Figure 8 As shown, step 103 includes:

[0173] Step 801, determine the weight corresponding to each dimension.

[0174] Step 802, based on the scores corresponding to the candidate government affairs items in each dimension and the weights corresponding to each dimension, weighted summation is performed to obtain a comprehensive score of the candidate government affairs items.

[0175] In an exemplary embodiment, the comprehensive score = (score on the data source dimension * weight 1) + (score on the relevance dimension * weight 2) + (score on the hit matter tag dimension * weight 3) + (score on the matter heat dimension * weight 4) + (score on the release time dimension * weight 5) + (score on the matter evaluation dimension * weight 6) + (and score on the handling frequency dimension * weight 7).

[0176] For example, the comprehensive score = (score on the data source dimension*2) + (score on the relevance dimension*3) + (score on the hit matter label dimension) + (score on the matter popularity dimension) + (score on the release time dimension) + (score on the matter evaluation dimension) + (score on the handling frequency dimension).

[0177] By comprehensively considering multiple dimensions such as data source dimension, relevance dimension, hit item label dimension, item popularity dimension, release time dimension, item evaluation dimension and handling frequency dimension, we ensure that search results can prioritize high-quality items that best match user needs.

[0178] It provides a set of flexible multi-dimensional sorting models that can be adjusted according to different government application scenarios to meet specific business needs. Whether it is the query needs for inquiring policies and regulations, business processes or convenient services, the system can achieve personalized sorting through different weight combinations. This flexible sorting method improves the applicability and scalability of the system.

[0179] In other embodiments, step 103 includes: calculating the average of the scores corresponding to the candidate government affairs items in various dimensions to obtain a comprehensive score of the candidate government affairs items.

[0180] Step 104, determining the ranking results of the candidate government affairs items based on the comprehensive scores of the candidate government affairs items.

[0181] In an exemplary embodiment, step 104 may include: sorting the candidate government affairs items in descending order according to their comprehensive scores to obtain a sorting result of the candidate government affairs items.

[0182] In an exemplary embodiment, when the comprehensive scores of multiple candidate government affairs are the same, the candidate government affairs are sorted in descending order according to the scores on the relevance dimension, to ensure that the user can obtain the most appropriate results.

[0183] It can effectively solve the multi-dimensional sorting problem in the search of government affairs, achieve accurate, timely and authoritative search result sorting, and improve the efficiency and satisfaction of users in searching for government affairs.

[0184] Step 105: Based on the ranking results of the candidate government affairs, determine the government affairs recommendation results corresponding to the question input by the user.

[0185] In an exemplary embodiment, step 105 may include: determining the first N candidate government affairs items in the ranking result of the candidate government affairs items in a sequence composed of the ranking order as the government affairs item recommendation result corresponding to the question input by the user.

[0186] For example, the first 50 candidate government affairs items in the ranking results of the candidate government affairs items are sequenced in the sorting order and determined as the government affairs item recommendation result corresponding to the question input by the user.

[0187] In an exemplary embodiment, step 105 may also include: directly determining the ranking results of candidate government affairs as the government affairs recommendation results corresponding to the user input question; step 105 may also include other schemes, which are not limited by the present application.

[0188] In summary, in this application, a search is performed based on the user input question to obtain candidate government affairs items that match the user input question, and the scores corresponding to the candidate government affairs items in each dimension are determined. The candidate government affairs items that match the user input question are not directly recommended to the user, but the scores corresponding to the candidate government affairs items in each dimension are determined, and the candidate government affairs items are scored through multiple dimensions. Then, based on the scores corresponding to the candidate government affairs items in each dimension, the comprehensive scores of the candidate government affairs items are determined. Compared with the traditional single keyword search, the comprehensive scores of the candidate government affairs items can better reflect the degree of match between the candidate government affairs items and the actual needs of the user. Based on the comprehensive scores of the candidate government affairs items, the ranking results of the candidate government affairs items are determined, and the candidate government affairs items are ranked based on the comprehensive scores of the candidate government affairs items. Then, based on the ranking results of the candidate government affairs items, the recommended results of the government affairs items corresponding to the user input question are determined. On the basis that the comprehensive scores of the candidate government affairs items can better reflect the degree of match between the candidate government affairs items and the actual needs of the user, the government affairs items recommendation results can be more accurately matched to the actual needs of the user by recommending after sorting, further improving the accuracy of government affairs item recommendations, and improving the efficiency and satisfaction of users in finding government affairs items.

[0189] Exemplary Devices

[0190] Correspondingly, the embodiment of the present application also provides a device for recommending government affairs, such as Fig. 9 As shown, the government affairs recommendation device includes:

[0191] A search unit 901 is used to search based on a question input by a user to obtain candidate government affairs items that match the question input by the user;

[0192] The first processing unit 902 is used to determine the scores corresponding to the candidate government affairs in each dimension;

[0193] The second processing unit 903 is used to determine the comprehensive score of the candidate government affairs item based on the scores corresponding to each dimension of the candidate government affairs item;

[0194] A ranking unit 904 is used to determine the ranking result of the candidate government affairs items based on the comprehensive scores of the candidate government affairs items;

[0195] The recommendation unit 905 is used to determine the government affairs item recommendation result corresponding to the user input question based on the ranking result of the candidate government affairs items.

[0196] Optionally, the dimension includes a data source dimension;

[0197] The first processing unit 902 is specifically configured to:

[0198] Determine a basic score of the data source of the candidate government affairs matter; wherein the basic score is positively correlated with the degree of correlation between the data source and the handling of the government affairs matter;

[0199] Determining a credibility score of the data source of the candidate government affairs matter;

[0200] Based on the basic score and the credibility score, the score of the candidate government affairs item in the data source dimension is determined.

[0201] Optionally, the dimension includes a relevance dimension; wherein the relevance dimension is used to characterize the relevance between the candidate government affairs item and the user input question;

[0202] The first processing unit 902 is specifically configured to:

[0203] Extracting each keyword in the question input by the user;

[0204] For any keyword in the question input by the user, perform the following operations:

[0205] Determining a first text match between the any one keyword and the name of the candidate government affairs item;

[0206] Determining a second text matching degree between the any one keyword and the specific content of the candidate government affairs item;

[0207] Determining a matching degree between the any one keyword and the candidate government affairs item based on the first text matching degree and the second text matching degree;

[0208] Based on the matching degree between each keyword in the question input by the user and the candidate government affairs, the score of the candidate government affairs in the relevance dimension is determined.

[0209] Optionally, the dimension includes a hit event label dimension;

[0210] The first processing unit 902 is specifically configured to:

[0211] Determine each target tag among the tags of the candidate government affairs matters that matches the question input by the user;

[0212] Based on each of the target tags, the score of the candidate government affairs item in the hit item tag dimension is determined.

[0213] Optionally, the dimension includes a matter popularity dimension;

[0214] The first processing unit 902 is specifically configured to:

[0215] At first preset time intervals, obtaining the query popularity of the candidate government affairs items;

[0216] Determining a popularity score of the candidate government affairs matter based on the query popularity;

[0217] At every second preset time, the heat score is decayed according to a preset decay ratio to obtain a heat score after decay;

[0218] Based on the attenuated heat score, the score of the candidate government affairs item in the item heat dimension is determined.

[0219] Optionally, the dimension includes a release time dimension;

[0220] The first processing unit 902 is specifically configured to:

[0221] Determining the release interval of the candidate government affairs item based on the release time of the candidate government affairs item and the current time;

[0222] Based on the release interval of the candidate government affairs, the score of the candidate government affairs in the release time dimension is determined.

[0223] Optionally, the dimension includes a matter evaluation dimension;

[0224] The first processing unit 902 is specifically configured to:

[0225] Based on the user service evaluation of the candidate government affairs item, the score of the candidate government affairs item in the item evaluation dimension is determined.

[0226] Optionally, the dimension includes a handling frequency dimension;

[0227] The first processing unit 902 is specifically configured to:

[0228] Based on the handling frequency of the candidate government affairs items within a preset time period, the score of the candidate government affairs items in the handling frequency dimension is determined.

[0229] Optionally, the dimensions include data source dimension, relevance dimension, hit item label dimension, item popularity dimension, release time dimension, item evaluation dimension and handling frequency dimension; wherein the relevance dimension is used to characterize the relevance between the candidate government affairs item and the user input question;

[0230] The corresponding scores in each dimension include the score in the data source dimension, the score in the relevance dimension, the score in the hit matter label dimension, the score in the matter popularity dimension, the score in the release time dimension, the score in the matter evaluation dimension, and the score in the handling frequency dimension.

[0231] Optionally, the second processing unit 903 is specifically configured to:

[0232] Determine the weights corresponding to each dimension;

[0233] Based on the scores corresponding to the candidate government affairs in each dimension and the weights corresponding to each dimension, a weighted sum is performed to obtain a comprehensive score of the candidate government affairs.

[0234] The government affairs recommendation device provided in this embodiment belongs to the same application concept as the government affairs recommendation method provided in the above embodiments of this application, and can execute the government affairs recommendation method provided in any of the above embodiments of this application, and has the corresponding functional modules and beneficial effects of executing the government affairs recommendation method. For technical details not fully described in this embodiment, please refer to the specific processing content of the government affairs recommendation method provided in the above embodiments of this application, and will not be repeated here.

[0235] The functions implemented by the above search unit 901, the first processing unit 902, the second processing unit 903, the sorting unit 904 and the recommendation unit 905 can be implemented by the same or different processors respectively, and the embodiment of the present application is not limited thereto.

[0236] It should be understood that the units in the above devices can be implemented in the form of a processor calling software. For example, the device includes a processor, the processor is connected to a memory, and instructions are stored in the memory. The processor calls the instructions stored in the memory to implement any of the above methods or realize the functions of each unit of the device, wherein the processor can be a general-purpose processor, such as a CPU or a microprocessor, etc., and the memory can be a memory in the device or a memory outside the device. Alternatively, the units in the device can be implemented in the form of hardware circuits, and the functions of some or all units can be realized by designing the hardware circuits. The hardware circuit can be understood as one or more processors; for example, in one implementation, the hardware circuit is an ASIC, and the functions of some or all of the above units are realized by designing the logical relationship of the components in the circuit; for another example, in another implementation, the hardware circuit can be implemented by PLD, taking FPGA as an example, which can include a large number of logic gate circuits, and the connection relationship between the logic gate circuits is configured by the configuration file, so as to realize the functions of some or all of the above units. All units of the above devices can be implemented in the form of a processor calling software, or in the form of hardware circuits, or in part by a processor calling software, and the remaining part is implemented in the form of hardware circuits.

[0237] In an embodiment of the present application, a processor is a circuit with the ability to process signals. In one implementation, the processor may be a circuit with the ability to read and run instructions, such as a CPU, a microprocessor, a GPU, or a DSP; in another implementation, the processor may implement certain functions through the logical relationship of a hardware circuit, and the logical relationship of the hardware circuit is fixed or reconfigurable, such as a hardware circuit implemented by an ASIC or PLD, such as an FPGA. In a reconfigurable hardware circuit, the process of the processor loading a configuration document to implement the hardware circuit configuration can be understood as the process of the processor loading instructions to implement the functions of some or all of the above units. In addition, it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as an NPU, TPU, DPU, etc.

[0238] It can be seen that each unit in the above device can be one or more processors (or processing circuits) configured to implement the above method, such as: CPU, GPU, NPU, TPU, DPU, microprocessor, DSP, ASIC, FPGA, or a combination of at least two of these processor forms.

[0239] In addition, all or part of the units in the above device can be integrated together, or can be implemented independently. In one implementation, these units are integrated together and implemented in the form of a SOC. The SOC may include at least one processor for implementing any of the above methods or implementing the functions of each unit of the device. The type of the at least one processor may be different, for example, including a CPU and an FPGA, a CPU and an artificial intelligence processor, a CPU and a GPU, etc.

[0240] Exemplary Electronic Devices

[0241] An embodiment of the present application provides an electronic device, see Fig.10 As shown, the device includes:

[0242] Memory 200 and processor 210;

[0243] The memory 200 is connected to the processor 210 and is used to store programs;

[0244] The processor 210 is used to implement the government affairs recommendation method disclosed in any of the above embodiments by running the program stored in the memory 200.

[0245] Specifically, the electronic device may further include: a bus, a communication interface 220 , an input device 230 and an output device 240 .

[0246] The processor 210, the memory 200, the communication interface 220, the input device 230 and the output device 240 are connected to each other via a bus.

[0247] A bus may include a pathway that transfers information between components of a computer system.

[0248] The processor 210 may be a general-purpose processor, such as a general-purpose central processing unit (CPU), a microprocessor, etc., or an application-specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of the program of the scheme of the present invention. It may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0249] The processor 210 may include a main processor, and may also include a baseband chip, a modem, and the like.

[0250] The memory 200 stores a program for executing the technical solution of the present invention, and may also store an operating system and other key services. Specifically, the program may include a program code, and the program code includes a computer operation instruction. More specifically, the memory 200 may include a read-only memory (ROM), other types of static storage devices that can store static information and instructions, a random access memory (RAM), other types of dynamic storage devices that can store information and instructions, a disk storage, a flash, and the like.

[0251] The input device 230 may include a device for receiving data and information input by a user, such as a keyboard, a mouse, a camera, a scanner, a light pen, a voice input device, a touch screen, a pedometer, or a gravity sensor.

[0252] Output device 240 may include a device that allows information to be output to a user, such as a display screen, a printer, a speaker, etc.

[0253] The communication interface 220 may include any device such as a transceiver to communicate with other devices or communication networks, such as Ethernet, a radio access network (RAN), a wireless local area network (WLAN), etc.

[0254] The processor 210 executes the program stored in the memory 200 and calls other devices, which can be used to implement each step of any government affairs recommendation method provided in the above embodiments of the present application.

[0255] Exemplary computer program products and storage media

[0256] In addition to the above-mentioned methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the government affairs recommendation method according to various embodiments of the present application described in any of the above embodiments of this specification.

[0257] The computer program product may be written in any combination of one or more programming languages ​​to write program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages, such as Java, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0258] In addition, the embodiment of the present application may also be a storage medium on which a computer program is stored. The computer program is executed by a processor to perform the steps of the government affairs recommendation method according to various embodiments of the present application described in any of the above embodiments of this specification, and specifically the following steps may be implemented:

[0259] Step 101, search based on the question input by the user to obtain candidate government affairs items that match the question input by the user.

[0260] Step 102, determining the scores corresponding to the candidate government affairs in each dimension.

[0261] Step 103: Determine the comprehensive score of the candidate government affairs item based on the scores of the candidate government affairs item in each dimension.

[0262] Step 104, determining the ranking results of the candidate government affairs items based on the comprehensive scores of the candidate government affairs items.

[0263] Step 105: Based on the ranking results of the candidate government affairs, determine the government affairs recommendation results corresponding to the question input by the user.

[0264] For the aforementioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0265] It should be noted that each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0266] The steps in the methods of each embodiment of the present application can be adjusted in sequence, combined and deleted according to actual needs, and the technical features recorded in each embodiment can be replaced or combined.

[0267] The modules and sub-modules in the devices and terminals in the various embodiments of the present application can be combined, divided and deleted according to actual needs.

[0268] In the several embodiments provided in the present application, it should be understood that the disclosed terminals, devices and methods can be implemented in other ways. For example, the terminal embodiments described above are only schematic, for example, the division of modules or submodules is only a logical function division, and there may be other division methods in actual implementation, for example, multiple submodules or modules can be combined or integrated into another module, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms.

[0269] The modules or submodules described as separate components may or may not be physically separated, and the components of the modules or submodules may or may not be physical modules or submodules, that is, they may be located in one place, or they may be distributed on multiple network modules or submodules. Some or all of the modules or submodules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0270] In addition, each functional module or submodule in each embodiment of the present application may be integrated into one processing module, or each module or submodule may exist physically separately, or two or more modules or submodules may be integrated into one module. The above-mentioned integrated modules or submodules may be implemented in the form of hardware or in the form of software functional modules or submodules.

[0271] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0272] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly by hardware, software units executed by a processor, or a combination of the two. The software units may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0273] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0274] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for recommending government affairs, characterized in that: include: Searching based on the question input by the user to obtain candidate government affairs items that match the question input by the user; Determine the scores corresponding to the candidate government affairs in each dimension; Determining a comprehensive score of the candidate government affairs item based on the scores corresponding to each dimension of the candidate government affairs item; Determining a ranking result of the candidate government affairs matters based on the comprehensive scores of the candidate government affairs matters; Based on the ranking results of the candidate government affairs, a government affairs matter recommendation result corresponding to the question input by the user is determined.

2. The government affairs recommendation method according to claim 1, characterized in that: The dimensions include data source dimensions; Determining the scores corresponding to the candidate government affairs in various dimensions includes: Determine a basic score of the data source of the candidate government affairs matter; wherein the basic score is positively correlated with the degree of correlation between the data source and the handling of the government affairs matter; Determining a credibility score of the data source of the candidate government affairs matter; Based on the basic score and the credibility score, the score of the candidate government affairs item in the data source dimension is determined.

3. The government affairs recommendation method according to claim 1, characterized in that: The dimension includes a relevance dimension; wherein the relevance dimension is used to characterize the relevance between the candidate government affairs item and the user input question; Determining the scores corresponding to the candidate government affairs in various dimensions includes: Extracting each keyword in the question input by the user; For any keyword in the question input by the user, perform the following operations: Determining a first text match between the any one keyword and the name of the candidate government affairs item; Determining a second text matching degree between the any one keyword and the specific content of the candidate government affairs item; Determining a matching degree between the any one keyword and the candidate government affairs item based on the first text matching degree and the second text matching degree; Based on the matching degree between each keyword in the question input by the user and the candidate government affairs, the score of the candidate government affairs in the relevance dimension is determined.

4. The government affairs recommendation method according to claim 1, characterized in that: The dimension includes a hit event label dimension; Determining the scores corresponding to the candidate government affairs in various dimensions includes: Determine each target tag among the tags of the candidate government affairs matters that matches the question input by the user; Based on each of the target tags, the score of the candidate government affairs item in the hit item tag dimension is determined.

5. The government affairs recommendation method according to claim 1, characterized in that: The dimensions include the matter popularity dimension; Determining the scores corresponding to the candidate government affairs in various dimensions includes: At first preset time intervals, obtaining the query popularity of the candidate government affairs items; Determining a popularity score of the candidate government affairs matter based on the query popularity; At every second preset time, the heat score is decayed according to a preset decay ratio to obtain a heat score after decay; Based on the attenuated heat score, the score of the candidate government affairs item in the item heat dimension is determined.

6. The government affairs recommendation method according to claim 1, characterized in that: The dimensions include a release time dimension; Determining the scores corresponding to the candidate government affairs in various dimensions includes: Determining the release interval of the candidate government affairs item based on the release time of the candidate government affairs item and the current time; Based on the release interval of the candidate government affairs, the score of the candidate government affairs in the release time dimension is determined.

7. The government affairs recommendation method according to claim 1, characterized in that: The dimensions include the matter evaluation dimension; Determining the scores corresponding to the candidate government affairs in various dimensions includes: Based on the user service evaluation of the candidate government affairs item, the score of the candidate government affairs item in the item evaluation dimension is determined.

8. The government affairs recommendation method according to claim 1, characterized in that: The dimensions include the dimension of handling frequency; Determining the scores corresponding to the candidate government affairs in various dimensions includes: Based on the handling frequency of the candidate government affairs items within a preset time period, the score of the candidate government affairs items in the handling frequency dimension is determined.

9. The government affairs recommendation method according to claim 1, characterized in that: The dimensions include data source dimension, relevance dimension, hit item label dimension, item popularity dimension, release time dimension, item evaluation dimension and handling frequency dimension; wherein the relevance dimension is used to characterize the relevance between the candidate government affairs item and the user input question; The corresponding scores in each dimension include the score in the data source dimension, the score in the relevance dimension, the score in the hit matter label dimension, the score in the matter popularity dimension, the score in the release time dimension, the score in the matter evaluation dimension, and the score in the handling frequency dimension.

10. The government affairs recommendation method according to any one of claims 1 to 9, characterized in that: Determining the comprehensive score of the candidate government affairs item based on the scores corresponding to each dimension of the candidate government affairs item includes: Determine the weights corresponding to each dimension; Based on the scores corresponding to the candidate government affairs in each dimension and the weights corresponding to each dimension, a weighted sum is performed to obtain a comprehensive score of the candidate government affairs.

11. A device for recommending government affairs, characterized in that: include: A search unit, configured to search based on a question input by a user, and obtain candidate government affairs items that match the question input by the user; A first processing unit is used to determine the scores corresponding to the candidate government affairs in each dimension; A second processing unit, configured to determine a comprehensive score of the candidate government affairs item based on the scores corresponding to the candidate government affairs items in each dimension; A ranking unit, configured to determine a ranking result of the candidate government affairs items based on the comprehensive scores of the candidate government affairs items; The recommendation unit is used to determine the government affairs item recommendation result corresponding to the user input question based on the ranking result of the candidate government affairs items.

12. An electronic device, characterized in that: including memory and processor; The memory is connected to the processor and is used to store programs; The processor is used to implement the government affairs recommendation method as described in any one of claims 1 to 10 by running the program in the memory.

13. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by the processor, the government affairs recommendation method according to any one of claims 1 to 10 is implemented.

14. A computer program product, characterized in that It comprises computer program instructions, which, when executed by a processor, enable the processor to execute the government affairs recommendation method as described in any one of claims 1 to 10.

Images