CN113779433A - Search result diversification and equalization searching method and computer equipment - Google Patents
Search result diversification and equalization searching method and computer equipment Download PDFInfo
- Publication number
- CN113779433A CN113779433A CN202110936147.1A CN202110936147A CN113779433A CN 113779433 A CN113779433 A CN 113779433A CN 202110936147 A CN202110936147 A CN 202110936147A CN 113779433 A CN113779433 A CN 113779433A
- Authority
- CN
- China
- Prior art keywords
- search
- preset data
- data model
- content
- equalization
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9538—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9532—Query formulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to a search result diversification and equalization searching method and computer equipment. The method comprises the following steps: s1, establishing an industry word bank, wherein the industry word bank comprises a plurality of industry professional vocabularies; converting each type of original data model into a preset data model; s2, receiving search contents input by a user, and extracting search keywords from the search contents according to an industry word bank; s3, retrieving all preset data models by using the search keywords, calculating the total weight value of each preset data model in the search results, and sorting the search results according to the total weight values. According to the invention, various types of original data models are uniformly converted into the preset data models, so that the influence on the search caused by the expression form of the data types is avoided, and the search results are diversified and balanced.
Description
Technical Field
The invention relates to the field of search, in particular to a search result diversification and equalization search method and computer equipment.
Background
The search technology is a common technology of the internet, and a user searches for target content by inputting search content. In the existing search technology, mostly only the relevance between the search content and the target content is considered, for example, the relevance is higher as the occurrence frequency is higher, the ranking display is performed according to the relevance, the variety diversity of the target content is not considered in the search mode, so that some kinds of target content are rarely displayed, some kinds of target content are too much displayed, and the search result is not diversified and balanced enough.
Disclosure of Invention
The present invention provides a search result diversification and equalization search method and a computer device, which are directed to the above-mentioned defects in the prior art.
The technical scheme adopted by the invention for solving the technical problems is as follows: a search result diversification equalization search method is constructed, and the method comprises the following steps:
s1, establishing an industry word bank, wherein the industry word bank comprises a plurality of industry professional vocabularies; converting each type of original data model into a preset data model;
s2, receiving search contents input by a user, and extracting search keywords from the search contents according to the industry word bank;
s3, retrieving all the preset data models by using the search keywords, calculating the total weight value of each preset data model in the search results, and sorting the search results according to the total weight values.
Further, in the search result diversification and equalization search method of the present invention, the preset data model includes a content title, a content abstract, a text, a keyword, and a content type.
Further, in the search result diversification and equalization search method according to the present invention, the step S1 of converting each type of original data model into a preset data model includes:
and converting each type of original data model into a preset data model and setting the weight value of each part of content of the preset data model, wherein the weight value of the keyword is greater than that of the content title, the weight value of the content title is greater than that of the content abstract, and the weight value of the content abstract is greater than that of the text.
Further, in the search result diversification and equalization search method according to the present invention, the step S3 of calculating the total weight value of each preset data model in the search result includes: and respectively calculating the partial weight values of the search keywords in the content title, the content abstract, the text, the keywords and the content type, and obtaining a total weight value from all the partial weight values.
Further, in the search result diversification equalization search method of the present invention, when calculating the partial weight values of the search keyword in the content title, the content abstract, the text, the keyword, and the content type, the partial weight values are positively correlated with the occurrence frequency of the search keyword.
Further, in the search result diversification equalization search method according to the present invention, after the step S3, the method further includes:
s4, adjusting the weight value of each part of the content of the preset data model according to the distribution condition of each type of data model corresponding to the preset data model in the search result, so that each type of data model in the search result is distributed evenly.
Further, in the search result diversification and equalization search method according to the present invention, the step S3 of retrieving all the preset data models by using the search keyword includes:
s31, classifying all the preset data models according to classification standards;
s32, counting the total number of the preset data types in each category, and dividing the categories with the same total number into the same group;
and S33, retrieving all the preset data models in each group by using the search keywords.
Further, in the search result diversification equalization search method according to the present invention, after the step S33, the method further includes: a preset number of preset data models are generated for each group.
Further, in the search result diversification and equalization search method of the present invention, the corresponding preset number of each group is positively correlated with the total number of the group.
In addition, the invention also provides computer equipment which comprises a memory and a processor, wherein the processor is in communication connection with the memory. The memory is used for storing a computer program; the processor is configured to execute the computer program stored in the memory to implement the search result diversification equalization search method as described above.
The search result diversification and equalization search method and the computer equipment have the following beneficial effects that: according to the invention, various types of original data models are uniformly converted into the preset data models, so that the influence on the search caused by the expression form of the data types is avoided, and the search results are diversified and balanced.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
fig. 1 is a flowchart of a search result diversification equalization search method according to an embodiment of the present invention;
fig. 2 is a flowchart of a search result diversification equalization search method according to an embodiment of the present invention.
Detailed Description
For a more clear understanding of the technical features, objects and effects of the present invention, embodiments of the present invention will now be described in detail with reference to the accompanying drawings.
In a preferred embodiment, referring to fig. 1, the search result diversification equalization search method of this embodiment includes the following steps:
s1, establishing an industry word bank, wherein the industry word bank comprises a plurality of industry professional vocabularies; and converting each type of original data model into a preset data model. Specifically, the term "professional" refers to a term used in a certain industry, and the term is a proper noun different from a daily life term and has a special meaning in the industry. The arrangement of the industry word bank is beneficial to scientific word segmentation of the search content input by the user, so that the search specialty and accuracy are improved. The industry word stock can be one or more, and when the industry word stock is multiple, the industry word stock is classified according to the content types to form multiple industry word stocks of different types, namely, the industry professional vocabularies contained in each industry word stock are the vocabularies of the same type.
In the prior art, the original data model is used during searching, namely, the original format of original data is kept for direct searching, and because various original data models are different, and various original data models are not on the same starting line, certain types in the searching result are displayed too much, certain types are displayed too little, and even certain types cannot be displayed all the time, so that the searching result is not diversified and balanced enough. For example, original data models such as news, movies, songs, encyclopedias, and anagories all contain "liu de hua", and because news, movies, songs, encyclopedias, and anagories have great differences in data types, when searching by using word frequency, news contains more "liu de hua" keywords, which results in that the news is basically news before the ranking in search results, and movies, songs, encyclopedias, and anagories are rarely found, especially movies and songs, and because of the limitation of data types, the news and the anagories are rarely found directly in the search results of the first page, so that the search results seen by users are single and not enough in diversified equalization. In order to solve the problem of insufficient diversification and equalization of search results caused by data model differences, the embodiment converts various types of original data models into preset data models, and all the converted original data models have uniform data models, so that all the preset data models have the same starting line, the probability that all the preset data models are searched when being retrieved is more balanced, and the search results are more diversified and equalized. Alternatively, the industry lexicon and all the converted preset data models are stored on the server.
And S2, receiving the search content input by the user, and extracting the search keywords from the search content according to the industry word bank. Specifically, a user inputs search content in a search box, the search content is uploaded to a server through a network, the server divides the search content according to industry professional vocabularies in an industry word bank, and search keywords corresponding to the search content are extracted. For example, the search content is "epson S1C17801 mcu data manual", and the word segmentation result is obtained by identifying according to an industry word stock: "epson" is a brand word, "S1C 17801" is a model word, "mcu" is a category word, and "data book" is a resource word, and then the extracted search keywords are: "epson", "S1C 17801", "mcu", "data sheet". Alternatively, if the industry thesaurus does not cover the search content, the search keyword can be extracted by using the basic language structure, namely, the main meaning object shape supplementary language structure is used for analyzing the search content to obtain the search keyword.
S3, retrieving all preset data models by using the search keywords, calculating the total weight value of each preset data model in the search results, and sorting the search results according to the total weight values. Specifically, if the search content only contains one search keyword, all preset data models are retrieved by using the search keyword, the total weight value of each preset data model in the search result is calculated, and the search result is sorted according to the total weight value. If the search content comprises at least two search keywords, firstly, using one search keyword to retrieve all preset data models to obtain a first search result; then, searching in the first search structure by using another search keyword to obtain a second search result; and repeating the steps until all the search keywords are searched, calculating the total weight value of each preset data model in the search results after the search is finished, sorting the search results according to the total weight values, and issuing the search results to the user terminal for display after the server finishes sorting. It can be understood that the search result sent by the server to the user terminal is not a preset data model, but an original data model corresponding to the preset data model.
According to the embodiment, the various types of original data models are uniformly converted into the preset data models, so that the influence on the search caused by the expression form of the data types is avoided, and the search results are diversified and balanced.
In some embodiments, the preset data model includes a content title, a content abstract, a text, a keyword, and a content type, and when each type of original data model is converted into the preset data model, the converted preset data model has the content title, the content abstract, the text, the keyword, and the content type regardless of whether the original data model has the content title, the content abstract, the text, the keyword, and the content type. For example, a song file usually has only the song title and artist information, and no content abstract and text, and the song lyrics can be used as the content abstract and the content text, so that the conversion is completed. According to the embodiment, the various types of original data models are uniformly converted into the preset data models, so that the influence on the search caused by the expression form of the data types is avoided, and the search results are diversified and balanced.
In the search result diversification equalization search method according to some embodiments, the converting, in step S1, each type of raw data model into a preset data model includes: and converting each type of original data model into a preset data model and setting the weight value of each part of content of the preset data model, wherein the weight value of the keyword is greater than that of the content title, the weight value of the content title is greater than that of the content abstract, and the weight value of the content abstract is greater than that of the text. Correspondingly, the step S3 of calculating the total weight value of each preset data model in the search result includes: and respectively calculating the partial weight values of the search keywords in the content title, the content abstract, the text, the keywords and the content type, and obtaining the total weight value from all the partial weight values. Alternatively, all the partial weight values are directly summed to obtain the total weight value. In addition, when calculating the partial weight values of the search keywords in the content title, the content abstract, the text, the keyword and the content type, the partial weight values are positively correlated with the occurrence times of the search keywords, that is, the more times the search keywords occur in a certain part, the larger the partial weight value obtained in the part. The difference between the original data models is balanced through weight configuration and the unified preset data models, so that the search results are more diversified and balanced.
In the search result diversification equalization search method according to some embodiments, with reference to fig. 2, after step S3, the method further includes: and S4, adjusting the weight value of each part of the content of the preset data model according to the distribution condition of each type of data model corresponding to the preset data model in the search result so as to balance the distribution of each type in the search result. The distribution condition of each type of data model corresponding to the preset data model in the search result means whether the preset data model corresponding to each type of data model appears in the preset ranking number (the search result displays the first page), and if the preset data model corresponding to each type of data model appears in the preset ranking number, the existing weight value is set reasonably; if the preset data models corresponding to one or more types of data models do not appear in the search results with the preset ranking number, which indicates that the existing weight values are unreasonably set and diversified and balanced search results cannot be realized, the weight values of the contents of all parts of the preset data models need to be adjusted to enable the distribution of all types in the search results to be balanced.
Further, the distribution condition of the preset data models corresponding to each type of data model in the search result means the proportion of the preset data models corresponding to each type of data model in the preset ranking number (the search result displays the first page), and if the proportion of the preset data models corresponding to each type of data model in the preset ranking number is balanced, the existing weight value setting is relatively reasonable; if the proportion of the preset data model corresponding to one or more types of data models in the preset ranking number is too low or too high, and diversified and balanced search results cannot be realized, the weight values of the contents of all parts of the preset data models need to be adjusted, so that the distribution of all types in the search results is balanced.
In the embodiment, the weight values of the contents of all the parts of the preset data model are adjusted through the feedback of the search result, and the weight value setting of the contents of all the parts of the preset data model is continuously optimized, so that the search result is more diversified and balanced.
In the search result diversification and equalization search method according to some embodiments, the step S3 of retrieving all the preset data models by using the search keyword includes:
and S31, classifying all the preset data models according to the classification standard. The classification standard can be flexibly selected according to the needs of users, such as a manufacturer class, a processor class, a resource class and the like, and further such as a news class, a movie class, a song class, an encyclopedia class, a hedonic class and the like.
And S32, counting the total number of the preset data types in each category, and dividing the categories with the equivalent total number into the same group, wherein the equivalent total number means that the total number is in the same preset number range. For example, there are some types of the predetermined data types with more than 1000 ten thousand, some types of the predetermined data types with between 500 ten thousand and 1000 ten thousand, some types of the predetermined data types with between 100 ten thousand and 500 ten thousand, some types of the predetermined data types with between 50 ten thousand and 100 ten thousand, some types of the predetermined data types with between 10 ten thousand and 50 ten thousand, some types of the predetermined data types with less than 10 ten thousand, and so on. Correspondingly, the total number of the category A and the category B is 650 ten thousand and 850 ten thousand respectively, and then the category A and the category B form a group; the total number of the category C and the category D is 65 ten thousand and 85 ten thousand respectively, and then the category C and the category D form a group; the total number of the category E and the category F is 6 ten thousand and 8 ten thousand, respectively, and the category E and the category F are one group.
S33, retrieving all preset data models in each group using the search keyword. Specifically, all the preset data models in each group are retrieved by using the search keywords respectively, and the search results of the search keywords in the group are obtained. In order to balance the distribution of each type in the search results, it is required to ensure that each group has a preset number of ranking numbers (the search results display the first page), and each group is required to generate a preset number of data models, and the corresponding preset number of each group is positively correlated with the total number of the group. That is, the more the total number of the group, the more the group occupies in the preset ranking number (the search result display first page), so that not only can each type be ensured to be displayed, but also the more the total number is, the more the preset data model is displayed, and the search results are more diversified and balanced.
The method and the device perform grouping according to the number level, retrieve in each group respectively, and ensure that each group has preset data model output, so that search results are more diversified and balanced.
In a preferred embodiment, the computer device of this embodiment includes a memory and a processor, the processor being communicatively coupled to the memory. The memory is used for storing a computer program; the processor is used for executing the computer program stored in the memory to realize the search result diversification equalization search method of the embodiment. Alternatively, the computer device is a server. The computer equipment of the embodiment uniformly converts various types of original data models into the preset data models, so that the influence on the search due to the expression form of the data types is avoided, and the search results are diversified and balanced.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The above embodiments are merely illustrative of the technical ideas and features of the present invention, and are intended to enable those skilled in the art to understand the contents of the present invention and implement the present invention, and not to limit the scope of the present invention. All equivalent changes and modifications made within the scope of the claims of the present invention should be covered by the claims of the present invention.
Claims (10)
1. A search result diversification equalization search method is characterized by comprising the following steps:
s1, establishing an industry word bank, wherein the industry word bank comprises a plurality of industry professional vocabularies; converting each type of original data model into a preset data model;
s2, receiving search contents input by a user, and extracting search keywords from the search contents according to the industry word bank;
s3, retrieving all the preset data models by using the search keywords, calculating the total weight value of each preset data model in the search results, and sorting the search results according to the total weight values.
2. The search result diversification and equalization search method according to claim 1, wherein the preset data model comprises a content title, a content abstract, a text, a keyword and a content type.
3. The search result diversification equalization search method according to claim 2, wherein the step S1 of converting each type of raw data model into a preset data model comprises:
and converting each type of original data model into a preset data model and setting the weight value of each part of content of the preset data model, wherein the weight value of the keyword is greater than that of the content title, the weight value of the content title is greater than that of the content abstract, and the weight value of the content abstract is greater than that of the text.
4. The search result diversification equalization search method according to claim 3, wherein the step S3 of calculating the total weight value of each preset data model in the search result comprises: and respectively calculating the partial weight values of the search keywords in the content title, the content abstract, the text, the keywords and the content type, and obtaining a total weight value from all the partial weight values.
5. The search result diversification and equalization search method according to claim 4, wherein when calculating the partial weight values of the search keyword in the content title, the content abstract, the text, the keyword and the content type, the partial weight values are positively correlated with the occurrence times of the search keyword.
6. The search result diversification equalization search method according to claim 3, further comprising, after the step S3:
s4, adjusting the weight value of each part of the content of the preset data model according to the distribution condition of each type of data model corresponding to the preset data model in the search result, so that each type of data model in the search result is distributed evenly.
7. The search result diversification and equalization search method according to claim 1, wherein the step S3 of retrieving all the preset data models by using the search keyword comprises:
s31, classifying all the preset data models according to classification standards;
s32, counting the total number of the preset data types in each category, and dividing the categories with the same total number into the same group;
and S33, retrieving all the preset data models in each group by using the search keywords.
8. The search result diversification equalization search method according to claim 7, further comprising, after the step S33: a preset number of preset data models are generated for each group.
9. The search result diversification equalization search method according to claim 8, wherein the preset number corresponding to each group is positively correlated to the total number of the groups.
10. A computer device comprising a memory and a processor, the processor communicatively coupled to the memory;
the memory is used for storing a computer program;
the processor is configured to execute the memory-stored computer program to implement the search result diversification equalization search method according to any one of claims 1 to 9.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110936147.1A CN113779433A (en) | 2021-08-16 | 2021-08-16 | Search result diversification and equalization searching method and computer equipment |
| PCT/CN2022/112863 WO2023020506A1 (en) | 2021-08-16 | 2022-08-16 | Search method with diversified and equalized search results, and computer device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110936147.1A CN113779433A (en) | 2021-08-16 | 2021-08-16 | Search result diversification and equalization searching method and computer equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113779433A true CN113779433A (en) | 2021-12-10 |
Family
ID=78837861
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110936147.1A Pending CN113779433A (en) | 2021-08-16 | 2021-08-16 | Search result diversification and equalization searching method and computer equipment |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN113779433A (en) |
| WO (1) | WO2023020506A1 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023020506A1 (en) * | 2021-08-16 | 2023-02-23 | 深圳市世强元件网络有限公司 | Search method with diversified and equalized search results, and computer device |
| CN116501841A (en) * | 2023-06-26 | 2023-07-28 | 深圳市唯特视科技有限公司 | Fuzzy query method, system and storage medium for data model |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105653661A (en) * | 2015-12-29 | 2016-06-08 | 云南电网有限责任公司电力科学研究院 | Search result re-ranking method and device |
| CN105938493A (en) * | 2016-04-14 | 2016-09-14 | 乐视控股（北京）有限公司 | Resource search method and apparatus |
| US20160328402A1 (en) * | 2015-05-06 | 2016-11-10 | App Annie Inc. | Keyword Reporting for Mobile Applications |
| CN111177532A (en) * | 2019-12-02 | 2020-05-19 | 平安资产管理有限责任公司 | Vertical search method, device, computer system and readable storage medium |
| CN111460104A (en) * | 2020-04-01 | 2020-07-28 | 神思电子技术股份有限公司 | Industry self-adaptive intelligent search method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103514299B (en) * | 2013-10-18 | 2018-04-17 | 北京奇虎科技有限公司 | Information search method and device |
| CN113779433A (en) * | 2021-08-16 | 2021-12-10 | 深圳市世强元件网络有限公司 | Search result diversification and equalization searching method and computer equipment |
-
2021
- 2021-08-16 CN CN202110936147.1A patent/CN113779433A/en active Pending
-
2022
- 2022-08-16 WO PCT/CN2022/112863 patent/WO2023020506A1/en unknown
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160328402A1 (en) * | 2015-05-06 | 2016-11-10 | App Annie Inc. | Keyword Reporting for Mobile Applications |
| CN105653661A (en) * | 2015-12-29 | 2016-06-08 | 云南电网有限责任公司电力科学研究院 | Search result re-ranking method and device |
| CN105938493A (en) * | 2016-04-14 | 2016-09-14 | 乐视控股（北京）有限公司 | Resource search method and apparatus |
| CN111177532A (en) * | 2019-12-02 | 2020-05-19 | 平安资产管理有限责任公司 | Vertical search method, device, computer system and readable storage medium |
| CN111460104A (en) * | 2020-04-01 | 2020-07-28 | 神思电子技术股份有限公司 | Industry self-adaptive intelligent search method |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023020506A1 (en) * | 2021-08-16 | 2023-02-23 | 深圳市世强元件网络有限公司 | Search method with diversified and equalized search results, and computer device |
| CN116501841A (en) * | 2023-06-26 | 2023-07-28 | 深圳市唯特视科技有限公司 | Fuzzy query method, system and storage medium for data model |
| CN116501841B (en) * | 2023-06-26 | 2023-09-08 | 深圳市唯特视科技有限公司 | Fuzzy query method, system and storage medium for data model |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2023020506A1 (en) | 2023-02-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108628971B (en) | Text classification method, text classifier and storage medium for unbalanced data set | |
| CN100423004C (en) | Video search dispatching system based on content | |
| CN107885745B (en) | Song recommendation method and device | |
| US20060155694A1 (en) | Query routing | |
| CN102163228B (en) | Method, apparatus and device for determining sorting result of resource candidates | |
| CN101419614A (en) | Video resource clustering method and device | |
| CN113779433A (en) | Search result diversification and equalization searching method and computer equipment | |
| CN102043843A (en) | Method and obtaining device for obtaining target entry based on target application | |
| CN110888990A (en) | Text recommendation method, device, equipment and medium | |
| CN106997390A (en) | A kind of equipment part or parts commodity transaction information search method | |
| CN109063171B (en) | Resource matching method based on semantics | |
| KR20010113779A (en) | Multi-feature combination generation and classification effectiveness evaluation using genetic algorithms | |
| CN111428007B (en) | Cross-platform based synchronous push feedback method | |
| CN112667866A (en) | Test paper generation method and device, electronic equipment and storage medium | |
| CN111488453B (en) | Resource grading method, device, equipment and storage medium | |
| CN108345694B (en) | Document retrieval method and system based on theme database | |
| CN117171331B (en) | Professional field information interaction method, device and equipment based on large language model | |
| CN112148999B (en) | Web API recommendation method based on Mashup service neighborhood under Web open environment | |
| US10474700B2 (en) | Robust stream filtering based on reference document | |
| JP2020035069A (en) | Database creation device and search system | |
| CN111104422B (en) | Training method, device, equipment and storage medium of data recommendation model | |
| CN115062135B (en) | Patent screening method and electronic equipment | |
| US9330170B2 (en) | Relating objects in different mediums | |
| CN105893397A (en) | Video recommendation method and apparatus | |
| CN112100330B (en) | Topic searching method and system based on artificial intelligence technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |