WO2023151576A1 - Procédé de recommandation de recherche, système de recommandation de recherche, dispositif informatique, et support de stockage - Google Patents

Procédé de recommandation de recherche, système de recommandation de recherche, dispositif informatique, et support de stockage Download PDF

Info

Publication number
WO2023151576A1
WO2023151576A1 PCT/CN2023/074947 CN2023074947W WO2023151576A1 WO 2023151576 A1 WO2023151576 A1 WO 2023151576A1 CN 2023074947 W CN2023074947 W CN 2023074947W WO 2023151576 A1 WO2023151576 A1 WO 2023151576A1
Authority
WO
WIPO (PCT)
Prior art keywords
search
user
recommendation
words
index
Prior art date
Application number
PCT/CN2023/074947
Other languages
English (en)
Chinese (zh)
Inventor
刘杨
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2023151576A1 publication Critical patent/WO2023151576A1/fr

Links

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
    • G06F16/9535—Search customisation based on user profiles and personalisation
    • 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
    • G06F16/906—Clustering; Classification
    • 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
    • G06F16/951—Indexing; Web crawling techniques
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Definitions

  • the present disclosure relates to the technical field of search, and in particular to a search recommendation method, a search recommendation system, computer equipment, and a storage medium.
  • the current search and sorting method is as follows: after the user enters a keyword as a search term, the background server of the system directly matches the search term with the search term in the knowledge base to obtain multiple search target articles. Then, for each article, the ranking score of each article is calculated according to the "relevance score", and these articles are sorted and recommended to users according to the score.
  • the correlation score represents the matching degree between the user's search term and the article, and is calculated by the system server using a specific algorithm.
  • the main purpose of the embodiments of the present disclosure is to provide a search recommendation method, a search recommendation system, a computer device, and a storage medium.
  • An embodiment of the present disclosure provides a search recommendation method, including: obtaining the index words corresponding to the search sentence input by the user; obtaining the user's search recommendation words, and the search recommendation words are used to associate the user's historical search behavior; according to the index words and search recommendation words Get search results.
  • An embodiment of the present disclosure also provides a search recommendation system, including: an index word acquisition module, configured to acquire an index word corresponding to a search sentence input by a user; a user preference module, configured to acquire a user's search recommendation word, and the search recommendation word is used associate The user's historical search behavior; the search module is used to obtain search results based on index words and search recommendation words.
  • An embodiment of the present disclosure also provides a computer device, the computer device includes a processor, a memory, a computer program stored on the memory and executable by the processor, and a data bus for realizing connection and communication between the processor and the memory, wherein When the computer program is executed by the processor, it realizes the steps of any search and recommendation method provided in the present disclosure.
  • An embodiment of the present disclosure also provides a storage medium for computer-readable storage.
  • the storage medium stores one or more programs, and one or more programs can be executed by one or more processors, so as to implement the information provided in this disclosure specification. Any one of the searches for recommended method steps.
  • FIG. 1 is a schematic flowchart of a search and recommendation method provided by an embodiment of the present disclosure
  • FIG. 2 is a schematic flowchart of a search and recommendation method in the telecommunications industry according to an embodiment of the present disclosure
  • FIG. 3 is a schematic diagram of a dictionary tree constructed in Embodiment 1 of the present disclosure.
  • FIG. 4 is a schematic structural diagram of a search recommendation system provided by an embodiment of the present disclosure.
  • FIG. 5 is a schematic structural block diagram of a computer device provided by an embodiment of the present disclosure.
  • Embodiments of the present disclosure provide a search recommendation method, system, computer equipment, and storage medium, which can improve the existing knowledge search that cannot be associated with the user's historical behavior and cannot accurately judge the user's search intention, resulting in insufficient accuracy of the search result. Improved user experience.
  • the search recommendation method can be applied to mobile terminals, and the mobile terminals may include electronic devices such as mobile phones, tablet computers, notebook computers, desktop computers, personal digital assistants, and wearable devices.
  • FIG. 1 is a schematic flowchart of a search and recommendation method provided by an embodiment of the present disclosure.
  • the search recommendation method includes steps S101 to S103.
  • Step S101 acquiring index words corresponding to a search sentence input by a user.
  • the search sentence input by the user is matched with a preset dictionary tree to obtain index words corresponding to the search sentence.
  • a preset dictionary tree By classifying the articles in the preset knowledge base, the preset dictionary tree is constructed according to the word segmentation result of the content of the article and the category to which the article belongs.
  • the business category corresponding to the index word can also be obtained according to the category in the preset dictionary tree.
  • the professional terms and articles in the telecommunications industry are collected first, and the articles are classified according to the business; the professional terms and common terms in the telecommunications industry are added to the thesaurus, and the contents of the knowledge articles are respectively subjected to coarse-grained word segmentation and fine-grained word segmentation.
  • Word segmentation results are obtained by deduplication.
  • Coarse-grained word segmentation means that when a sentence contains custom common words in the telecommunications industry, word segmentation is performed according to custom words.
  • knowledge articles can be segmented by an ANSJ tokenizer.
  • Step S102 acquiring the user's search recommendation words, which are used to correlate the user's historical search behavior.
  • the search recommendation words preset by the user may be obtained from the user database according to the user ID of the user, and the search recommendation words are used together with the index words matched with the search sentence for searching. Furthermore, in each search, the user's historical search behavior can be associated. On the one hand, the input information of the search is expanded, and on the other hand, it is more based on the user's search recommendation words to make the search results closer to the user's search intent.
  • the user's historical search and article reading behavior can be analyzed based on the neural network, and then the user's search recommendation words can be updated and maintained according to the analysis results .
  • the network model analyzes the user's information and search history to obtain the user's search preference, and updates the user's search recommendation words according to the search preference.
  • Step S103 obtaining search results according to the index words and the search recommendation words.
  • the articles in the knowledge base are searched according to the index words and the search recommendation words.
  • the required search results are obtained by filtering the articles in the knowledge base that do not contain the index words and the search recommendation words.
  • the search results can also be sorted, so that the search results that are more relevant to the user's search intention are displayed in the front, so that the user can locate the required knowledge articles more quickly .
  • a correlation score is performed on the search results, and the search results are sorted according to the obtained correlation scores to obtain recommended results.
  • the search results need to be sorted, the articles that better meet the user's search intention can be sorted in a higher position. It is necessary to score the relevance of the search results. According to the scoring results, articles with higher scores are closer to the user's search intent. Therefore, the search results are sorted in order from high to low, and the sorted search results are obtained, that is, the recommended results displayed to the user.
  • the relevance score of the search results is performed by setting different weights for index words, synonyms, and search recommendation words.
  • the index word is a search entity word matched according to the user's search statement, so the highest weight is set, and the synonym of the index word is set with the second highest weight, and the user's search recommendation word represents the user's historical search Behavior, may have little connection with this search, so set the third highest weight.
  • TF-IDF Term Frequency-Inverse Document Frequency, term frequency-inverse document frequency
  • the recommendation result is optimized according to the preference classification to obtain the optimized recommendation result; wherein, the optimization of the recommendation result according to the preference classification includes: weighting the relevance scores of the search results belonging to the preference classification, and according to The obtained weighted relevance scores rank the search results to obtain optimized recommendation results.
  • the user's preference classification can be set by the user in advance according to their own business needs. In order to maintain the user's preference classification more intelligently, it can also intelligently analyze the user's historical behavior and information of various dimensions based on the neural network, so as to intelligently Ability to maintain and update user preference categories.
  • collecting user information includes: collecting user information in various dimensions, such as user position information, customer-oriented types, and working hours; user search history data includes: user's browsing order of articles, number of times read articles, read article time, whether the article is a favorite, etc., and the above-mentioned search history data of the user are used as neurons in the input layer of the neural network.
  • x i is the i-th input neuron data
  • w i is the weight of the i-th input neuron data
  • b i is the offset of the hidden layer.
  • the user's preference score for the most recently read articles is obtained, and the articles with high preference scores for the recently read articles are counted, and the user preference classification score is calculated according to the category of the article, the preference score, and the number of classified articles according to a self-defined algorithm. Value, classify the category with high score as the user's preference.
  • the user's access frequency can also be recorded for the top-ranked articles. If the user's access frequency of the article is very small, or the user does not access it, it will be considered that such an article does not meet the user's search intention. , and perform a subtraction operation on it when sorting.
  • category analysis may be performed on articles that are rarely accessed by users to identify categories that users dislike, and use the categories that users dislike as a basis for reducing points when sorting search results.
  • Exemplarily record the access frequency of the first 100 articles recommended to the user, by associating the ID of the article in the database with the user ID, count the number of visits, and obtain the number of visits when sorting, the number of visits is very When it is small or 0, the corresponding article will be deducted.
  • the categories of these low-frequency access articles are obtained.
  • the category is considered to be the user's non-preferred category.
  • search results subtract points for search results belonging to non-preferred categories.
  • the scores of the articles in the search results are sorted from high to low, and the sorted search results are returned to the user as recommended results.
  • the words are used to score the relevance of the articles and sort them according to the scoring results, so that the articles that are more in line with the user's search intention are displayed in the front, which is convenient for the user to find the desired article more quickly.
  • the search recommendation method extracts the index words contained in the search sentence based on the dictionary tree of the knowledge base, and obtains the user's search recommendation words and/or search preferences for searching, and expands the search sentence to obtain More results that match user search intent.
  • the search recommendation words and preference categories representing the user's search preferences are used together with the search index words to score the relevance of the articles and sort them according to the scoring results, so that the articles that are more in line with the user's search intention are displayed first.
  • the user's search preference is predicted and analyzed through deep learning of user search and browsing data, and the analysis result is used to modify the user's search preference, so that the user's search preference can be intelligently adjusted according to the search history.
  • the search recommendation results that are more in line with the user's intention are obtained, and the user's satisfaction is improved.
  • the technical solutions of the embodiments of the present disclosure improve the search results by associating the user's historical search preferences when searching, and optimize the ranking of the search results, so that the user's search intention can be judged more accurately, and the user's experience is improved.
  • embodiments of the present disclosure also provide two specific embodiments on the search recommendation method, which are as follows.
  • FIG. 2 is a schematic flowchart of a search and recommendation method in the telecommunications industry according to an embodiment of the present disclosure
  • FIG. 3 is a schematic diagram of a dictionary tree constructed in Embodiment 1. This embodiment specifies search recommendations in specific scenarios only for simplicity, and the present disclosure is also applicable to search recommendations in other scenarios.
  • Step 1 Collect knowledge articles in the telecommunications industry, for example: “Internet Domestic Data Transmission”, “Internet Data Center”, “Call Center”. And the article is classified according to the content: “Internet domestic data transmission” belongs to the basic telecommunications business, “Internet data center” and “call center” belong to the value-added telecommunications business.
  • Step 2 Carry out word segmentation for knowledge articles.
  • the word segmentation results for "internet domestic data transmission” include: “Internet”, “domestic", “data”, and “transmission”
  • the word segmentation results for "Internet Data Application Center” include: “ Internet”, “data”, “center”, “data center”
  • the word segmentation results for "call center” include: “call", "center”.
  • Table 1 shows the results of word segmentation for the above knowledge articles.
  • Step 3 obtain the user ID (such as Test1), and obtain the search sentence entered by the user, such as "Internet data”.
  • Step 4 the user's search sentence is matched through the nodes of the dictionary tree, and the index words "Internet” and “data” are obtained, and the matching categories are "basic telecommunication services” and "value-added telecommunication services”.
  • Step 5 obtain the search recommendation words in the user search preferences stored in the database through the user ID, such as "domestic".
  • Step 6 Perform a matching search in different classification indexes according to the category of the index words and search recommendation words, and the matched articles include: "Internet domestic data transmission" and "Internet data center”.
  • Step 7 Set different weight values for index words and search recommended words, where the weight of index words is greater than the weight of search recommended words, and use the TF-IDF scoring algorithm to score matching articles according to the weights of index words and search recommended words.
  • the TF-IDF model calculates a value for each document D and the query string Q composed of keywords W[1]...W[k] according to TF and IDF, which is used to indicate the matching degree between the query string Q and the document D :
  • the weighting coefficient H is added, and the final query matching degree is:
  • the predetermined preference score ratio is shown in Table 4.
  • the database is queried through the user ID to obtain preference categories such as "basic telecommunication services", and then according to the preference category "basic telecommunication services", the articles belonging to the preference category in the search results are added points, and the "internet domestic Data Transfer" for further bonus points.
  • Step 8 Finally, sort the search results according to the scores from large to small to form a recommended result list, and finally return the recommended result list to the user.
  • Embodiment 2 of the present disclosure provides an application of a user preference screening method in a telecommunication customer service system, and the present disclosure is also applicable to preference screening in other application scenarios.
  • Step 1 collect user information in various dimensions, such as user position information (business consultation, fee inquiry, business handling, new business promotion, etc.), customer-oriented types (family customers, government and enterprise customers, public phones, wireless local calls), working hours, And the user's search and browsing data of articles, including: browsing order, number of times of reading articles, time of reading articles, whether to save or not, etc., input the user's various dimension information and search and browsing data to the neurons of the input layer.
  • user position information business consultation, fee inquiry, business handling, new business promotion, etc.
  • customer-oriented types family customers, government and enterprise customers, public phones, wireless local calls
  • working hours And the user's search and browsing data of articles, including: browsing order, number of times of reading articles, time of reading articles, whether to save or not, etc.
  • Step 2 Establish a BP neural network model for each category, use the above-mentioned information data of each dimension as the input of the BP neural network, use the activation function in the hidden layer to calculate and output the preference score for reading articles in the latest period, and obtain user information based on the BP neural network structure Recently read article preference score.
  • Step 3 Count the recently read articles with high preference scores, calculate the user preference classification score according to the category, preference score, and number of classified articles according to the article category, preference score, and the number of classified articles, and use the category with high score as the user's preference category.
  • Step 4 Count the articles with high preference scores of the recently read articles, analyze the coarse-grained word segmentation words with high frequency in each article according to the recently read articles, use such words as the user's search recommendation words, and correct and update the search in the user database Recommended words.
  • Step 5 continue to collect data on the user's reading behavior by using the ranking items pushed to the user, including: matching the number of clicks on the article, the length of reading, whether to save it, etc. as the input of step 1.
  • the top browsing order, multiple clicks, long reading time, and bookmarking an article reflect that the user likes the article; otherwise, it means that the user does not like the article.
  • the user's search recommendation words and preference classification are updated based on the user's historical behavior and user information based on the BP neural network.
  • FIG. 4 is a schematic diagram of a scene implementing the search recommendation system provided by this embodiment. As shown in FIG. Search module 203 .
  • the index word obtaining module 201 includes: a search request obtaining module 2011 and an index word matching module 2012 .
  • the search request obtaining module 2011 is used to receive the user's search statement
  • the index word matching module 2012 is used to extract the index words and classifications in the input search sentence.
  • the specific method for generating index words and classifications is as follows: by matching the search sentence with a preset dictionary tree, extracting the index words contained in the search sentence and the classification information corresponding to the index words from the dictionary tree.
  • the user preference acquisition module 202 includes a search recommendation word acquisition module 2021 and a preference classification acquisition module 2022.
  • the search recommendation word acquisition module 2021 is used to obtain the user's search recommendation words from the database according to the ID information of the user; the preference classification acquisition module 2022 uses The user's preference classification is obtained from the database according to the user's ID information, and the preference classification is used as one of the basis for subsequent ranking of search results.
  • the search module 203 includes a synonym acquisition module 2031 and a matching search module 2032 .
  • the synonym acquisition module 2031 is used to obtain the synonym of the index word extracted from the search sentence; the matching search module 2032 is used to And the user's search recommendation words are searched and matched from the knowledge database to obtain search results.
  • the search recommendation system further includes: a ranking module 204 , which can be specifically divided into: a weight marking module 2041 , a correlation scoring module 2042 , and a ranking recommendation module 2043 .
  • the weight marking module 2041 is used to assign different weights to index words, synonyms, and user's search recommendation words
  • the correlation scoring module 2042 is used to perform search results according to index words, synonyms, search recommendation words and corresponding different weights. Relevance score, the result of which will be used to sort the search results.
  • the sorting and recommending module 2043 is used to sort the search results in order of scores from high to low according to the relevance scoring results of the search results and the user's preference classification to form recommended results, and return the searched recommended results to the user.
  • the search recommendation system further includes: a user preference screening module 205 , specifically: a user information acquisition module 2051 , a browsing history acquisition module 2052 , a user preference analysis module 2053 , and a user preference correction module 2054 .
  • the user information acquisition module 2051 is used to collect the basic information of the user
  • the browsing history acquisition module 2052 is used to obtain the user's article search and historical data of the article (such as the order of browsing articles, the number of times they are read, and whether they are favorites)
  • the user preference analysis module 2053 is used to predict the degree of user preference for articles based on user information and reading history of articles combined with BP neural network, extract articles with high preference, summarize and fuse the number of preferred articles and degree of preference, and extract preference classification.
  • the search recommendation words are screened out by predetermined screening principles
  • the user preference modification module 2054 is used to update the preference classification and search recommendation words in the user database according to the analysis and prediction results of the BP neural network.
  • the search recommendation system further includes: a knowledge base classification and indexing module 206 , which can be specifically divided into: a knowledge base classification module 2061 , a knowledge base word segmentation module 2062 , and a dictionary tree construction module 2063 .
  • the knowledge base classification module 2061 is used to classify the articles in the knowledge base according to the business;
  • the knowledge base word segmentation module 2062 is used to extract the professional vocabulary and other commonly used custom vocabulary in the knowledge base articles as index words;
  • the dictionary tree construction sub-module 2063 It is used to construct a dictionary tree according to the classification of the indexed articles.
  • FIG. 5 is a schematic structural block diagram of a computer device provided by an embodiment of the present disclosure.
  • a computer device 300 includes a processor 301 and a memory 302, and the processor 301 and the memory 302 are connected through a bus 303, such as an I2C (Inter-integrated Circuit) bus.
  • a bus 303 such as an I2C (Inter-integrated Circuit) bus.
  • the processor 301 is used to provide computing and control capabilities to support the operation of the entire computer device.
  • the processor 301 can be a central processing unit (Central Processing Unit, CPU), and the processor 301 can also be other general-purpose processors, a digital signal processor (Digital Signal Processor, DSP), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC ), Field-Programmable Gate Array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • the general-purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
  • the memory 302 may be a Flash chip, a read-only memory (ROM, Read-Only Memory) disk, an optical disk, a U disk, or a mobile hard disk.
  • FIG. 5 is only a block diagram of a partial structure related to the embodiment of the present disclosure, and does not constitute a limitation on the computer equipment to which the embodiment of the present disclosure is applied.
  • the computer device may include more or fewer components than shown in the figures, or combine certain components, or have a different arrangement of components.
  • the processor is configured to run a computer program stored in the memory, and implement any search and recommendation method provided by the embodiments of the present disclosure when executing the computer program.
  • the processor is configured to run a computer program stored in the memory, and implement the following steps when executing the computer program: obtain the index word corresponding to the search sentence input by the user; obtain the user's search recommended word, and search the recommended word with Based on the historical search behavior of associated users; obtain search results based on index words and search recommendation words.
  • the processor when implementing the search recommendation method, is configured to: perform correlation scoring on search results, and sort the search results according to the obtained correlation scores to obtain recommendation results.
  • the processor when implementing the search and recommendation method, is configured to: obtain user preference classifications, optimize the recommendation results according to the preference classifications, and obtain optimized recommendation results; where the recommendation results are optimized according to the preference classifications
  • the optimization includes: weighting the relevance scores of the search results belonging to the preference category, and sorting the search results according to the obtained weighted relevance scores to obtain optimized recommendation results.
  • the processor when the processor implements the search recommendation method, it is used to: collect user information and search history, analyze the user information and search history based on the preset BP neural network model to obtain the user's search preference; Preferences update the user's search recommendation words and preference categories.
  • the processor when the processor obtains the index words corresponding to the search sentences input by the user, it is used to implement: perform word segmentation and matching on the search sentences input by the user and the preset dictionary tree, and obtain the index words corresponding to the search sentences .
  • the processor when implementing the search and recommendation method, is used to: classify the articles in the preset knowledge base, construct a preset dictionary according to the results of word segmentation of the content of the articles and the categories to which the articles belong Tree.
  • the processor when the processor acquires the search results according to the index words and search recommended words, it is used to realize: searching for synonyms associated with the index words from a preset relational database; Get search results.
  • the processor when the processor implements the correlation scoring of the search results, it is used to: set different weights for index words, synonyms and search recommendation words to perform correlation scoring on the search results, and obtain the search results relevance score.
  • the embodiment of the present disclosure also provides a storage medium for computer-readable storage, the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors, so as to realize the Steps for any one of the search recommended methods provided in the instructions.
  • the storage medium may be an internal storage unit of the computer device in the foregoing embodiments, such as a hard disk or memory of the computer device.
  • the storage medium can also be an external storage device of the computer equipment, such as a plug-in hard disk equipped on the computer equipment, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash memory card (Flash Card) wait.
  • a smart memory card Smart Media Card, SMC
  • SD Secure Digital
  • flash memory card Flash Card
  • the functional modules/units in the system, and the device can be implemented as software, firmware, hardware, and an appropriate combination thereof.
  • the division between functional modules/units mentioned in the above description does not necessarily correspond to the division of physical components; for example, one physical component may have multiple functions, or one function or step may be composed of several physical components. Components cooperate to execute.
  • Some or all of the physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit .
  • Such software may be distributed on computer readable media, which may include computer storage media (or non-transitory media) and communication media (or transitory media).
  • computer storage media includes both volatile and nonvolatile media implemented in any method or technology for storage of information, such as computer readable instructions, data structures, program modules, or other data. permanent, removable and non-removable media.
  • Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cartridges, tape, magnetic disk storage or other magnetic storage devices, or can Any other medium used to store desired information and which can be accessed by a computer.
  • communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media .

Abstract

Les modes de réalisation de la présente divulgation concernent un procédé de recommandation de recherche, un système de recommandation de recherche, un dispositif informatique, et un support de stockage, appartenant au domaine technique de la recherche. Le procédé consiste à : acquérir un terme d'index correspondant à une phrase de recherche entrée par un utilisateur ; obtenir un terme de recherche recommandé pour l'utilisateur, le terme de recherche recommandé étant utilisé pour se lier à un comportement de recherche historique de l'utilisateur ; et, en fonction du terme d'index et du terme de recherche recommandé, obtenir un résultat de recherche.
PCT/CN2023/074947 2022-02-08 2023-02-08 Procédé de recommandation de recherche, système de recommandation de recherche, dispositif informatique, et support de stockage WO2023151576A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210118904.9 2022-02-08
CN202210118904.9A CN116610853A (zh) 2022-02-08 2022-02-08 搜索推荐方法、搜索推荐系统、计算机设备及存储介质

Publications (1)

Publication Number Publication Date
WO2023151576A1 true WO2023151576A1 (fr) 2023-08-17

Family

ID=87563626

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/074947 WO2023151576A1 (fr) 2022-02-08 2023-02-08 Procédé de recommandation de recherche, système de recommandation de recherche, dispositif informatique, et support de stockage

Country Status (2)

Country Link
CN (1) CN116610853A (fr)
WO (1) WO2023151576A1 (fr)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117391824B (zh) * 2023-12-11 2024-04-12 深圳须弥云图空间科技有限公司 基于大语言模型和搜索引擎推荐物品的方法及装置

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110188186A (zh) * 2019-04-24 2019-08-30 平安科技（深圳）有限公司 医疗领域的内容推荐方法、电子装置、设备及存储介质
CN111737574A (zh) * 2020-06-19 2020-10-02 口口相传(北京)网络技术有限公司 搜索信息获取方法、装置、计算机设备及可读存储介质
KR20210011102A (ko) * 2019-07-22 2021-02-01 주식회사 앱컴파니 맞춤형 상업정보 인공지능 검색 시스템
CN113282832A (zh) * 2021-06-10 2021-08-20 北京爱奇艺科技有限公司 一种搜索信息的推荐方法、装置、电子设备及存储介质
CN113343091A (zh) * 2021-06-22 2021-09-03 力合科创集团有限公司 面向产业和企业的科技服务推荐计算方法、介质及程序

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110188186A (zh) * 2019-04-24 2019-08-30 平安科技（深圳）有限公司 医疗领域的内容推荐方法、电子装置、设备及存储介质
KR20210011102A (ko) * 2019-07-22 2021-02-01 주식회사 앱컴파니 맞춤형 상업정보 인공지능 검색 시스템
CN111737574A (zh) * 2020-06-19 2020-10-02 口口相传(北京)网络技术有限公司 搜索信息获取方法、装置、计算机设备及可读存储介质
CN113282832A (zh) * 2021-06-10 2021-08-20 北京爱奇艺科技有限公司 一种搜索信息的推荐方法、装置、电子设备及存储介质
CN113343091A (zh) * 2021-06-22 2021-09-03 力合科创集团有限公司 面向产业和企业的科技服务推荐计算方法、介质及程序

Also Published As

Publication number Publication date
CN116610853A (zh) 2023-08-18

Similar Documents

Publication Publication Date Title
US11663254B2 (en) System and engine for seeded clustering of news events
WO2019214245A1 (fr) Procédé et appareil de transfert d'informations, et dispositif terminal et support de stockage
US7756868B2 (en) Method for providing search results list based on importance information and system thereof
JP4920023B2 (ja) オブジェクト間競合指標計算方法およびシステム
WO2021098648A1 (fr) Procédé, appareil et dispositif de recommandation de texte, et support
US20160034514A1 (en) Providing search results based on an identified user interest and relevance matching
CN109299383B (zh) 生成推荐词的方法、装置、电子设备及存储介质
US9720979B2 (en) Method and system of identifying relevant content snippets that include additional information
WO2006108069A2 (fr) Recherche de contenu accessible par l'intermediaire de formulaires web
Im et al. Linked tag: image annotation using semantic relationships between image tags
WO2012129149A2 (fr) Regroupement de résultats de recherche basé sur l'association d'instances de données à des entités de bases de connaissances
KR20150016973A (ko) 탐색 결과들을 생성하는 방법
CN110390094B (zh) 对文档进行分类的方法、电子设备和计算机程序产品
WO2018176913A1 (fr) Procédé et appareil de recherche, et support d'informations lisible par ordinateur non temporaire
KR20180097120A (ko) 전자 문서 검색 방법 및 그 서버
CA2956627A1 (fr) Systeme et moteur servant au regroupement cible d'evenements d'informations
WO2023151576A1 (fr) Procédé de recommandation de recherche, système de recommandation de recherche, dispositif informatique, et support de stockage
AU2018313274B2 (en) Diversity evaluation in genealogy search
CN111930949B (zh) 搜索串处理方法、装置、计算机可读介质及电子设备
US9336280B2 (en) Method for entity-driven alerts based on disambiguated features
CN115630144A (zh) 一种文档搜索方法、装置及相关设备
Selvan et al. ASE: Automatic search engine for dynamic information retrieval
CN116610782B (zh) 文本检索方法、装置、电子设备及介质
CN112860940B (zh) 基于描述逻辑知识库上有序概念空间的音乐资源检索方法
KR101137491B1 (ko) 웹 페이지 검색에서 개인화된 태그 추천 모델 활용 시스템 및 방법

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23752356

Country of ref document: EP

Kind code of ref document: A1