Knowledge base interfacing method, apparatus, device, and medium

Abstract

The application relates to a knowledge base docking method, device, equipment and medium. The method comprises the following steps: constructing a knowledge base comprising multiple industries, wherein the multiple industries comprise a beauty industry and a mobile phone industry; detecting a current application scene, wherein the application scene comprises a search box application scene, a word cloud application scene, a content analysis application scene, an algorithm recommendation application scene and a chart drilling application scene; and docking the knowledge base by using a docking strategy matched with the application scene. The application solves the problem that a user cannot drill down for analysis while quickly and accurately obtaining a query result by docking the knowledge base with a strategy matched with a demand scene under different demand scenes.

Description

Technical Field

[0001] This application relates to the field of deep learning technology, and in particular to a knowledge base docking method, apparatus, device and medium. Background Technology

[0002] A knowledge base is a structured, easy-to-operate, easy-to-use, comprehensive, and organized cluster of knowledge in knowledge engineering. It is a collection of interconnected knowledge fragments stored, organized, managed, and used in computer memory using certain knowledge representation methods to meet the needs of solving problems in specific domains. Knowledge bases can be used to easily create, organize, search, and share knowledge, centralizing knowledge scattered across various locations. For example, a knowledge base can act as a portal to the knowledge of all members within an enterprise, playing a crucial role in the enterprise's knowledge management process. However, existing knowledge bases cannot simultaneously provide fast and accurate query results while meeting users' needs for drill-down analysis. Summary of the Invention

[0003] This application provides a knowledge base docking method, apparatus, equipment, and medium to solve the aforementioned technical problem that "existing knowledge bases cannot meet the user's drill-down analysis needs while quickly and accurately obtaining query results".

[0004] According to one aspect of the embodiments of this application, this application provides a knowledge base docking method, including: constructing a knowledge base including multiple industries, wherein the multiple industries include the beauty industry and the mobile phone industry; detecting the current application scenario, wherein the application scenario includes a search box application scenario, a word cloud application scenario, a content analysis application scenario, an algorithm recommendation application scenario, and a chart drill-down application scenario; and docking with the knowledge base using a docking strategy that matches the application scenario.

[0005] Optionally, the docking strategy that matches the application scenario with the knowledge base includes: in the case of a search box application scenario, obtaining the search field entered by the user; obtaining the keyword tags corresponding to the search field from the knowledge base; and querying the data corresponding to the search field from the knowledge base using the keyword tags as an index.

[0006] Optionally, obtaining keyword tags corresponding to the search field from the knowledge base includes: obtaining search history and determining a first tag that matches the search field from the search history; finding a second tag that matches the search field from the knowledge base; and integrating the first tag and the second tag to obtain keyword tags.

[0007] Optionally, the docking strategy matching the application scenario and docking with the knowledge base includes: in the case of a word cloud application scenario, obtaining the search field input by the user; obtaining the tag words that are related to the search field according to the knowledge base, and obtaining the first tag dimension of the tag words; counting the number of the first dimension of the first tag, and when the number of the first dimension is less than the first preset number, selecting the second tag dimension from the posts that mention the search field according to the volume, so that the sum of the number of the first dimension and the number of the second dimension of the second tag is the first preset number.

[0008] Optionally, the docking strategy that matches the application scenario with the knowledge base includes: when the application scenario is a content analysis application scenario, obtaining the search fields input by the user; obtaining the entity type corresponding to the search field according to the knowledge base, and displaying the entity type as a first-level tag; obtaining the entity tag corresponding to the entity type, and displaying the entity tag as a second-level tag when the first-level tag is selected.

[0009] Optionally, the docking strategy that matches the application scenario with the knowledge base includes: when the application scenario is an algorithm recommendation application scenario, obtaining the search field input by the user; obtaining the entity type corresponding to the search field according to the knowledge base, and obtaining a second preset number of entity tags corresponding to the entity type; extracting recommendation data related to the search field from the knowledge base, and displaying the recommendation data according to the entity tags.

[0010] Optionally, the docking strategy matching the application scenario and docking with the knowledge base includes: in the case of a chart drill-down application scenario, if a user-input search field is detected, docking with the industry database corresponding to the search field in the knowledge base; generating a word cloud corresponding to the search field, and if any word cloud is selected, performing content analysis on the selected word cloud according to the industry database; displaying the analysis results of the content analysis, and if any original post component in the analysis results is triggered, jumping to the original post data corresponding to the original post component.

[0011] According to another aspect of the embodiments of this application, this application also provides a knowledge base docking device, including: a construction module for constructing a knowledge base including multiple industries, wherein the multiple industries include the beauty industry and the mobile phone industry; a detection module for detecting the current application scenario, wherein the application scenario includes a search box application scenario, a word cloud application scenario, a content analysis application scenario, an algorithm recommendation application scenario, and a chart drill-down application scenario; and a docking module for docking with the knowledge base using a docking strategy matching the application scenario.

[0012] According to another aspect of the embodiments of this application, this application also provides an electronic device, including a memory, a processor, a communication interface and a communication bus. The memory stores a computer program that can run on the processor. The memory and the processor communicate with each other through the communication bus and the communication interface. When the processor executes the computer program, it implements the steps of any of the above methods.

[0013] According to another aspect of the embodiments of this application, this application also provides a computer-readable medium having processor-executable non-volatile program code that causes the processor to perform any of the methods described above.

[0014] The technical solution of this application can be applied to the design of natural language processing using deep learning technology.

[0015] Compared with related technologies, the technical solutions provided in this application have the following advantages:

[0016] This application provides a knowledge base integration method, comprising: constructing a knowledge base encompassing multiple industries, including the beauty industry and the mobile phone industry; detecting the current application scenario, including search box application scenario, word cloud application scenario, content analysis application scenario, algorithm recommendation application scenario, and chart drill-down application scenario; and integrating with the knowledge base using an integration strategy matched to the application scenario. This application solves the problem of not being able to quickly and accurately obtain query results while simultaneously meeting the user's drill-down analysis needs by integrating the knowledge base with a strategy matching the specific requirements of different scenarios. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart of an optional knowledge base integration method provided according to an embodiment of this application;

[0020] Figure 2 This is a block diagram of an optional knowledge base docking device provided according to an embodiment of this application;

[0021] Figure 3 This is a schematic diagram of an optional electronic device structure provided according to an embodiment of this application. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0023] In the following description, the use of suffixes such as "module," "part," or "unit" to denote elements is solely for the purpose of illustration and has no specific meaning in itself. Therefore, "module" and "part" may be used interchangeably.

[0024] A knowledge base is a structured, easy-to-operate, easy-to-use, comprehensive, and organized cluster of knowledge in knowledge engineering. It is a collection of interconnected knowledge fragments stored, organized, managed, and used in computer memory using certain knowledge representation methods to meet the needs of solving problems in specific domains. Knowledge bases can be used to easily create, organize, search, and share knowledge, centralizing knowledge scattered across various locations. For example, a knowledge base can act as a portal to the knowledge of all members within an enterprise, playing a crucial role in the enterprise's knowledge management process. However, existing knowledge bases cannot simultaneously provide fast and accurate query results while meeting users' needs for drill-down analysis.

[0025] To address the problems mentioned in the background art, according to one aspect of an embodiment of this application, this application provides a knowledge base integration method, such as... Figure 1 As shown, it includes:

[0026] Step 101: Construct a knowledge base that includes multiple industries, including the beauty industry and the mobile phone industry;

[0027] Step 103: Detect the current application scenario, which includes search box application scenario, word cloud application scenario, content analysis application scenario, algorithm recommendation application scenario, and chart drill-down application scenario.

[0028] Step 105: Connect with the knowledge base using a connection strategy that matches the application scenario.

[0029] This application relates to a knowledge base access method, and more particularly to an industry knowledge base access method, which connects a beauty industry knowledge base and a mobile phone industry knowledge base to an existing knowledge base. Then, according to different application scenarios, the corresponding access method is executed to connect with the knowledge base including the beauty industry knowledge base and the mobile phone industry knowledge base, thereby achieving accurate query results in different scenarios while meeting the user's drill-down analysis needs.

[0030] The tags in the industry knowledge base and the rules (including word logic) in the existing system are essentially two parallel query logics, representing two different forms of generating the analyzed objects, and thus yielding different results.

[0031] As an optional implementation, the docking strategy that matches the application scenario with the knowledge base includes: in the case of a search box application scenario, obtaining the search field entered by the user; obtaining the keyword tags corresponding to the search field from the knowledge base; and querying the data corresponding to the search field from the knowledge base using the keyword tags as an index.

[0032] Search count quotas are not deducted when users search for keyword tags; they are only deducted after selecting a tag and clicking search.

[0033] When a user searches for multiple tags, starting from the second tag, the search results can only be retrieved within the same industry knowledge base. This means that users are not allowed to compare tags from different industries. In other words, if it is the same knowledge base, searching and querying are allowed.

[0034] As an optional embodiment, obtaining keyword tags corresponding to the search field from the knowledge base includes: obtaining search history and determining a first tag that matches the search field from the search history; finding a second tag that matches the search field from the knowledge base; and integrating the first tag and the second tag to obtain keyword tags.

[0035] Keyword tags are search suggestions, mainly obtained through precise tag matching in the small database and search history. Ultimately, only four types of results are returned: keywords, tags, custom, and groups. The priority order from high to low is: keywords, tags, custom, and groups.

[0036] Optionally, when users search for keyword tags, they need to match the corresponding results in all the connected industry databases. If the keyword is matched in multiple knowledge bases, multiple results need to be returned, and the different results are distinguished by the naming of the industry database.

[0037] Optionally, users can perform "real-time" "exact matching" when searching for keyword tags.

[0038] Optionally, once a user selects a keyword tag, it cannot be edited.

[0039] Optionally, after the user selects a tag, the time component needs to restrict the selectable time to no less than a preset time (e.g., September 1, 2021).

[0040] Specifically, when users search for keywords in the search box, the analysis objects with tags in the history also need to be tagged with the most recently searched tags (such as custom tags, knowledge base tags, etc.). Hot topic lists, KOL quick search lists, and industry think tanks also retain the most recently searched tags when users search for keywords and the search history is hit.

[0041] Prioritize displaying searches with recent search tags. Searches without recent search tags will be displayed in the order of keyword-tag-custom-group. Since there are no groups in the search history, all searches will have recent search tags, and the combined results will still be displayed in the order of keyword-tag-custom.

[0042] As an optional implementation, the docking strategy that matches the application scenario and docks with the knowledge base includes: in the case of a word cloud application scenario, obtaining the search field input by the user; obtaining the tag words that are related to the search field according to the knowledge base, and obtaining the first tag dimension of the tag words; counting the number of the first dimension of the first tag, and when the number of the first dimension is less than a first preset number, selecting the second tag dimension from the posts that mention the search field according to the volume, so that the sum of the number of the first dimension and the number of the second dimension of the second tag is the first preset number.

[0043] First, query all posts based on tags and classify them according to "tags in the relationship". Obtain aggregated data for different tag dimensions and aggregate the tag count2 within the tag dimension. If the obtained tag dimension of the relationship is less than the first preset number (e.g., 8), obtain the post dimension from the mention tags of the hit post to make up 8 dimensions.

[0044] The word cloud in this application is a word cloud divided by entity type (including a general word cloud). When a user searches for a tag, all the original entity types and their corresponding entities are found in the posts that match the tag. Since only 3 entity types in the 3C knowledge base are generated by the algorithm, there are only 3 original entities in the 3C knowledge base.

[0045] Specifically, the entity type corresponding to the user's search tag also needs to be included in the statistics, but the original entity words of the search are not displayed under that entity type. Filtering is required based on the alignment table, and all alignment words appearing in the alignment table need to be filtered out. In the general word cloud, words for the name of the searched analysis object are also not displayed.

[0046] Entity types for which the original post contains no data will not be displayed. If no data is found for any entity type, the general term will be displayed by default. The entity types to be displayed will be determined by the following logic: When a user searches for a tag: First, retrieve the entity types corresponding to the original entities that have been related to that tag, up to 8, using volume as the statistical indicator (if multiple original entities in a post are related, the volume of the entity type will only be counted as 1); If no related original entities are retrieved, display the entity types corresponding to other original entities (i.e., mentioning other tags at the same time), up to 8, using volume as the statistical indicator (if multiple tags are mentioned at the same time in a post, the volume of the entity type will only be counted as 1); If fewer than 8 entity types corresponding to the original entities that have been related to that tag are retrieved, the entity types corresponding to other original entities will be selected in order of volume to make up the difference; The entity type corresponding to the user's search tag itself will also be included in the statistics, but the original entity searched for will not be displayed under that entity type; the name of the original entity searched for will also not be displayed in the general term cloud.

[0047] For example, if a user searches for "skincare products" and the posts that match the "skincare products" tag include the tag "L'Oréal," and L'Oréal corresponds to the entity type "brand," then "brand" is the entity type that has a relationship with "skincare products." However, if the tag mentioned at the same time as "skincare products" is "niacinamide," then "niacinamide" is a different entity type.

[0048] Specifically, the volume of a single entity type will not exceed the volume of the search object. The word cloud displayed under each entity type dimension consists of entity words (the entity words displayed here are "raw entity words"), and a maximum of the top 100 (volume statistics) entity words will be displayed; if multiple raw entities appear in a post (including related and simultaneous mentions), they need to be calculated separately when counting entity words; the volume of a single entity word will not exceed the volume of the entity type.

[0049] When a user searches for multiple tags, the entity type displayed will be based on this product; if a competitor's product cannot find a corresponding tag for this product's entity type, "NO DATA" will be displayed.

[0050] Specifically, the entity word cloud in this application supports hiding keywords, displaying metrics with mouse Hower, and drilling down to view the original post; it does not display the toggle button for volume and word frequency; it supports comparing up to two entity word clouds at the same time; it supports users to switch between different entity types by pulling down; it enumerates general entities plus entity dimensions; and it displays the entity type with the highest volume by default.

[0051] Next to the word cloud, there is an annotation button with a prompt that could be "Combined with NLP (Natural Language Processing) algorithm models, this displays the multi-dimensional results mentioned by consumers when discussing the search object under the current conditions."

[0052] As an optional implementation, the docking strategy that matches the application scenario with the knowledge base includes: when the application scenario is a content analysis application scenario, obtaining the search fields input by the user; obtaining the entity type corresponding to the search fields according to the knowledge base, and displaying the entity type as a first-level tag; obtaining the entity tag corresponding to the entity type, and displaying the entity tag as a second-level tag when the first-level tag is selected.

[0053] After the initial query for data statistics in the small database, the number of results needs to be validated. If there are fewer than 8 results, a second query is needed to supplement them to 8. The first query counts the "tag relationship" field, while the second query counts the "entity tag" field. The logic of the second query is quite different from the first query, so they are performed in two separate queries.

[0054] Specifically, the content analysis module is divided into a first level and a second level. The first level is the entity type, and the second level is the tags under the entity type.

[0055] It should be noted that the entity type corresponding to the user's search tag itself is not included in the statistical display. The first level is the entity type, and the statistical indicator is the volume, which is arranged in descending order of volume.

[0056] The displayed secondary tags are the entity tags under the corresponding primary entity type, and the statistical metric is volume. Only the top 10 tags are displayed at the secondary level. When a user searches for multiple tags, the entity type displayed is based on this product; if the corresponding competitor's tag cannot be found in this product's entity type, "NO DATA" will be displayed.

[0057] Users can expand to the second level by clicking the first-level tab; the first-level tab is selected by default. The first-level mouse cursor displays the logical logic of "the values ​​of all analyzed objects under a certain entity type"; the second-level mouse cursor displays the logical logic of "the values ​​of all analyzed objects under a certain entity tag". Clicking the second level allows users to drill down into the original post content for analysis. There is a note button next to it with the prompt text "Combining NLP algorithm models, aggregating information mentioned by consumers when discussing search objects under the current conditions and displaying key information based on the volume of voice".

[0058] As an optional embodiment, the docking strategy that matches the application scenario with the knowledge base includes: when the application scenario is an algorithm recommendation application scenario, obtaining the search field input by the user; obtaining the entity type corresponding to the search field according to the knowledge base, and obtaining a second preset number of entity tags corresponding to the entity type; extracting recommendation data related to the search field from the knowledge base, and displaying the recommendation data according to the entity tags.

[0059] In algorithm-based recommendation scenarios, only tag queries are performed. This application queries tag entity type statistics by recommending relevant tags through the algorithm, while existing technologies only query statistics for both tags and entity words. The two query logics differ, therefore they need to be processed separately.

[0060] Specifically, the recommended data consists of the performance of all entity tags in the entity type corresponding to the user's current search tag under the current filtering conditions. Since the user's search tag may overlap with the recommended tag, deduplication is required. Therefore, the number of recommendations by the algorithm follows the principle of "the number of recommendations after deduplication is at most 9" (based on the statistics of the current indicators).

[0061] For example, if a user searches for "Little Brown Bottle," and the entity type corresponding to "Little Brown Bottle" is "product," and the user's query range is data from Xiaohongshu from January to March, then the entities displayed in the recommendation list are the top 10 products that appeared most frequently in Xiaohongshu from January to March, sorted in descending order according to the current sorting criteria.

[0062] Among the recommended entities, entity tags that are mentioned simultaneously with the current search tag need to be marked.

[0063] For example, if a user searches for "Little Brown Bottle" and there are entity tags such as "Little Blue Bottle" and "Little Black Bottle" in the corresponding product entity type, and "Little Blue Bottle" is mentioned at the same time as "Little Brown Bottle", then it needs to be marked as "mentioned at the same time" in the list.

[0064] When users switch sorting criteria, new recommendation data needs to be recalculated and generated based on the new metrics; the supported sorting metrics are volume, interaction volume, and exposure volume.

[0065] Regarding the number of recommendations and the number of analysis objects (if the maximum number of analysis objects is 10): When the number of analysis objects in the search box is greater than 10, clicking on an algorithm recommendation will prompt "The current number of analysis objects is greater than 10, please adjust and try again"; when the number of analysis objects in the search box is less than 10, if the number of algorithm recommendations clicked by the user is greater than 10, the content in the search box will be displayed first, and the remaining items will be supplemented to 10 through algorithm recommendations; if the number of algorithm recommendations clicked by the user is less than 10, all items will be displayed.

[0066] Specifically, in algorithm recommendation applications, the overview displays the following metrics: volume, interaction, exposure, and sentiment index (sentiment refers to the entity's own sentiment, not the post's sentiment; if the searched entity hasn't been sentiment-tested, it's indicated by "-"). The overview defaults to displaying the top 3; clicking expands to see more. If a product's data isn't in the top 3, it should be placed fourth in the default unexpanded state; after expanding, its position follows the actual ranking. During rule queries, the list display logic follows the latest logic. When a user searches for multiple tags, considering a product's ranking might be lower, its ranking is displayed as follows: ① 0-100 (inclusive) directly displays the ranking; ② 100+ displays. Recommended entities need to be marked "Algorithm Recommended." The user's search and current status must be consistent. When the mouse hovers over the "Recommended" tag, if the entity also mentions this product, it displays "This product is mentioned in the data of this analysis object." The recommendation list displays the following metrics: volume, interaction, exposure, and sentiment index.

[0067] Additionally, when a user opens the algorithm recommendation, the entity tags recommended by the algorithm need to be automatically populated into the existing chart. There is a prompt button next to the algorithm recommendation; the prompt text is: "Based on the NLP algorithm model, automatically recommend the market performance of other analysis objects of the same type as the search object and automatically populate the chart."

[0068] As an optional implementation, the docking strategy that matches the application scenario with the knowledge base includes: in the case of a chart drill-down application scenario, if a user-input search field is detected, docking with the industry database corresponding to the search field in the knowledge base; generating a word cloud corresponding to the search field, and if any word cloud is selected, performing content analysis on the selected word cloud according to the industry database; displaying the analysis results of the content analysis, and if any original post component in the analysis results is triggered, jumping to the original post data corresponding to the original post component.

[0069] When drilling down a tag-type chart, query the original post in the corresponding industry knowledge base, including: switching the data source of Elasticsearch; replacing the QueryString in the DSL (Domain Specific Language) with tag query; and drilling down in word cloud and content analysis scenarios, supporting both tag dimension and tag drill-down.

[0070] The display logic for chart drill-down in this application is divided into word cloud, content analysis, and original post.

[0071] For example, a pop-up window appears after the user drills down; clicking on the second level of word cloud or content analysis will display the original post, which expands to the left.

[0072] Specifically, this application also relates to the merging of similar articles, including:

[0073] (1) The function to merge similar articles is not enabled by default;

[0074] (2) When a user clicks to merge duplicate posts, similar articles should be merged for the maximum of 100 original posts displayed on the current page; if any posts are merged, the page will display "Similar articles mentioning the keyword have been merged, totaling ++ posts"; if there are no duplicate posts, the page will display "No similar articles mentioning the keyword were found" (e.g., if a maximum of 100 posts are retrieved based on the most recently published criteria, and 20 of them are similar, these 20 similar posts should be merged together).

[0075] (3) Description of display order: When posts are merged, the displayed posts are displayed normally according to the current status (i.e., the corresponding sorting conditions); when some posts are merged, the displayed posts are first executed within the group according to the current sorting conditions, and then executed between the groups. Among them, "within the group" means a group of posts that are merged together, and one post is selected according to the current sorting conditions. "between the groups" means multiple groups of posts that are merged, and one post between multiple groups is displayed in the system according to the current sorting conditions.

[0076] (4) The number of posts displayed in the pagination is displayed in groups. For example, if 100 posts are aggregated into 10 groups, then all of them need to be displayed.

[0077] (5) Add a "similar article identifier" field to the exported data. If there are no similar articles, use N / A. If there are similar articles, use Arabic numerals. Similar articles grouped together need to be identified by the same numerical code.

[0078] By default, only one post is displayed when they are merged together. Users can click to expand to show all posts and click to collapse to return to the state of displaying only one post.

[0079] The above methods can be applied to real-time public opinion: public opinion trends, platform distribution, sentiment distribution, and user profiles; competitive product analysis: competitive product trends, competitive product platform distribution, simultaneous mentions of competitive products, and volume and sentiment distribution; and activity analysis: activity performance trends, activity analysis by platform, activity sentiment performance, user profiles, and activity KOL (Key Opinion Leader) performance.

[0080] This application provides a knowledge base integration method, comprising: constructing a knowledge base encompassing multiple industries, including the beauty industry and the mobile phone industry; detecting the current application scenario, including search box application scenario, word cloud application scenario, content analysis application scenario, algorithm recommendation application scenario, and chart drill-down application scenario; and integrating with the knowledge base using an integration strategy matched to the application scenario. This application solves the problem of not being able to quickly and accurately obtain query results while simultaneously meeting the user's drill-down analysis needs by integrating the knowledge base with a strategy matching the specific requirements of different scenarios.

[0081] According to another aspect of the embodiments of this application, this application also provides a knowledge base docking device, such as... Figure 2 As shown, it includes:

[0082] Module 202 is used to build a knowledge base that includes multiple industries, including the beauty industry and the mobile phone industry.

[0083] The detection module 204 is used to detect the current application scenario, which includes search box application scenario, word cloud application scenario, content analysis application scenario, algorithm recommendation application scenario and chart drill-down application scenario;

[0084] The docking module 206 is used to dock with the knowledge base using a docking strategy that matches the application scenario.

[0085] It should be noted that the construction module 202 in this embodiment can be used to execute step 101 in this application embodiment, the detection module 204 in this embodiment can be used to execute step 103 in this application embodiment, and the docking module 206 in this embodiment can be used to execute step 105 in this application embodiment.

[0086] Optionally, the docking module 206 further includes a first docking submodule, which is used to obtain the search field input by the user when the application scenario is a search box application scenario; obtain the keyword tags corresponding to the search field according to the knowledge base; and query the data corresponding to the search field from the knowledge base using the keyword tags as indexes.

[0087] Optionally, the first docking submodule is also used to obtain search history and determine the first tag that matches the search field from the search history; find the second tag that matches the search field from the knowledge base; and integrate the first tag and the second tag to obtain the keyword tag.

[0088] Optionally, the docking module 206 further includes a second docking submodule, used to obtain the search field input by the user in the case of a word cloud application scenario; obtain the tag words that are related to the search field according to the knowledge base, and obtain the first tag dimension of the tag words; count the number of the first dimension of the first tag, and when the number of the first dimension is less than the first preset number, select the second tag dimension from the posts that mention the search field according to the volume, so that the sum of the number of the first dimension and the number of the second dimension of the second tag is the first preset number.

[0089] Optionally, the docking module 206 also includes a third docking submodule, which is used to obtain the search fields input by the user when the application scenario is a content analysis application scenario; obtain the entity type corresponding to the search field according to the knowledge base, and display the entity type as a first-level tag; obtain the entity tag corresponding to the entity type, and display the entity tag as a second-level tag when the first-level tag is selected.

[0090] Optionally, the docking module 206 further includes a fourth docking submodule, which is used to obtain the search field input by the user when the application scenario is an algorithm recommendation application scenario; obtain the entity type corresponding to the search field according to the knowledge base, and obtain a second preset number of entity tags corresponding to the entity type; extract recommendation data related to the search field from the knowledge base, and display the recommendation data according to the entity tags.

[0091] Optionally, the docking module 206 also includes a fifth docking submodule, which, in the case of a chart drill-down application scenario, if a user-input search field is detected, docks with the industry database corresponding to the search field in the knowledge base; generates a word cloud corresponding to the search field; and, if any word cloud is selected, performs content analysis on the selected word cloud based on the industry database; displays the analysis results of the content analysis; and, if any original post component in the analysis results is triggered, jumps to the original post data corresponding to the original post component.

[0092] It should be noted that the examples and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the content disclosed in the above embodiments.

[0093] According to another aspect of the embodiments of this application, such as Figure 3 As shown, this application provides an electronic device, including a memory 31, a processor 32, a communication interface 33, and a communication bus 34. The memory 31 stores a computer program that can run on the processor 32. The memory 31 and the processor 32 communicate through the communication bus 34 and the communication interface 33. When the processor 32 executes the computer program, it implements the steps of the above method.

[0094] The memory and processor in the aforementioned electronic devices communicate with each other via a communication bus and a communication interface. The communication bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into an address bus, a data bus, a control bus, etc.

[0095] The memory may include random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0096] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0097] According to another aspect of the embodiments of this application, this application provides a computer-readable medium having processor-executable non-volatile program code that causes the processor to perform the steps of any of the methods described above.

[0098] Optionally, in embodiments of this application, the computer-readable medium is configured to store program code for the processor to perform the above-described method steps.

[0099] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments, and will not be repeated here.

[0100] In specific implementation, the embodiments of this application can be referred to the above embodiments and have corresponding technical effects.

[0101] It is understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, microcode, or a combination thereof. For hardware implementation, the processing unit can be implemented in one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontrollers, microprocessors, other electronic units for performing the functions described herein, or combinations thereof.

[0102] For software implementation, the techniques described herein can be implemented by units that perform the functions described herein. The software code can be stored in memory and executed by a processor. The memory can be implemented in the processor or external to the processor.

[0103] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0104] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0105] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0106] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0107] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0108] If the aforementioned function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiments of this application, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks. It should be noted that in this document, relational terms such as "first" and "second" are used only 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. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. In the absence of further restrictions, an element defined by the phrase "comprising a..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0109] The above description is merely a specific embodiment of this application, enabling those skilled in the art to understand or implement this application. Various modifications to these embodiments will be readily 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 this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A knowledge base integration method, characterized in that, include: Construct a knowledge base encompassing multiple industries, including the beauty industry and the mobile phone industry; Detect the current application scenarios, including search box application scenarios, word cloud application scenarios, content analysis application scenarios, algorithm recommendation application scenarios, and chart drill-down application scenarios; A docking strategy matching the application scenario is adopted to dock with the knowledge base; In the case where the application scenario is the word cloud application scenario, the connection strategy includes: obtaining the search field input by the user; obtaining the tag words that are related to the search field according to the knowledge base, and obtaining the first tag dimension of the tag words; counting the first dimension quantity of the first tag dimension; and when the first dimension quantity is less than a first preset quantity, selecting the second tag dimension from the posts that mention the search field according to the volume, so that the sum of the first dimension quantity and the second dimension quantity of the second tag dimension is the first preset quantity; wherein, when counting the volume, if there are multiple original entities related to the search field in a post, the volume of the first tag dimension or the second tag dimension is only counted as 1.

2. The method according to claim 1, characterized in that, The method of using a connection strategy that matches the application scenario to connect with the knowledge base includes: When the application scenario is the search box application scenario, obtain the search field input by the user; Obtain keyword tags corresponding to the search fields based on the knowledge base; The keyword tags are used as indexes to query data corresponding to the search field from the knowledge base.

3. The method according to claim 2, characterized in that, The step of obtaining the keyword tags corresponding to the search field based on the knowledge base includes: Retrieve the search history and determine the first tag that matches the search field from the search history; Search the knowledge base for a second tag that matches the search field; By integrating the first tag and the second tag, the keyword tag is obtained.

4. The method according to claim 1, characterized in that, The method of using a connection strategy that matches the application scenario to connect with the knowledge base includes: When the application scenario is the content analysis application scenario, the search fields input by the user are obtained; The entity type corresponding to the search field is obtained from the knowledge base, and the entity type is displayed as a first-level tag; Obtain the entity tag corresponding to the entity type, and when the first-level tag is selected, display the entity tag as a second-level tag.

5. The method according to claim 1, characterized in that, The method of using a connection strategy that matches the application scenario to connect with the knowledge base includes: When the application scenario is the application scenario recommended by the algorithm, the search fields input by the user are obtained; Based on the knowledge base, obtain the entity type corresponding to the search field, and obtain a second preset number of entity tags corresponding to the entity type; Recommended data related to the search field is extracted from the knowledge base, and the recommended data is displayed according to the entity tags.

6. The method according to claim 1, characterized in that, The method of using a connection strategy that matches the application scenario to connect with the knowledge base includes: In the case where the application scenario is the chart drill-down application scenario, if a search field entered by the user is detected, the system will connect to the industry database in the knowledge base that corresponds to the search field. Generate a word cloud corresponding to the search field, and if any word cloud is detected to be selected, perform content analysis on the selected word cloud according to the industry database; The analysis results of the content analysis are displayed, and if any original post component in the analysis results is triggered, the user is redirected to the original post data corresponding to the original post component.

7. A knowledge base docking device, characterized in that, include: The building module is used to build a knowledge base that includes multiple industries, including the beauty industry and the mobile phone industry. The detection module is used to detect the current application scenario, which includes search box application scenario, word cloud application scenario, content analysis application scenario, algorithm recommendation application scenario, and chart drill-down application scenario. The docking module is used to dock with the knowledge base using a docking strategy that matches the application scenario. Where the application scenario is the word cloud application scenario, the docking strategy includes: obtaining the user-input search field; obtaining tags related to the search field from the knowledge base, and obtaining the first tag dimension of the tags; counting the first dimension quantity of the first tag dimension; and when the first dimension quantity is less than a first preset quantity, selecting a second tag dimension from posts mentioning the search field based on the volume, so that the sum of the first dimension quantity and the second dimension quantity of the second tag dimension is the first preset quantity; wherein, when counting the volume, if a post contains multiple original entities associated with the search field, the volume of the first tag dimension or the second tag dimension is only counted as 1.

8. An electronic device comprising a memory, a processor, a communication interface, and a communication bus, wherein the memory stores a computer program executable on the processor, and the memory and the processor communicate via the communication bus and the communication interface, characterized in that... When the processor executes the computer program, it implements the steps of the method described in any one of claims 1 to 6.

9. A computer-readable medium having processor-executable non-volatile program code, characterized in that, The program code causes the processor to execute the method of any one of claims 1 to 6.

Images