Search processing methods, apparatus, electronic devices, computer-readable storage media, and computer program products

Abstract

This application provides a search processing method, apparatus, electronic device, computer-readable storage medium, and computer program product. The method includes: displaying query information entered through the information input operation in response to an information input operation; displaying multiple prompt words corresponding to the query information, wherein each prompt word is a separate option; and displaying search results corresponding to at least one prompt word selected by the selection operation in response to a selection operation. This application enables accurate identification of search intent, thereby improving the accuracy of search results.

Description

Technical Field

[0001] This application relates to computer technology, and more particularly to a search processing method, apparatus, electronic device, computer-readable storage medium, and computer program product. Background Technology

[0002] When users search for content in unfamiliar areas, they often struggle to accurately describe their search intent. This is because, on the one hand, users lack sufficient knowledge of the content they are querying, making it difficult for them to use standardized vocabulary to describe their search; on the other hand, users lack a systematic understanding of the target field and cannot ask key questions. Consequently, search engines receive too little information to clearly identify the user's true search intent. To find the answer, users must gradually approach the desired result by searching for similar terms, browsing content, and changing their search terms, impacting search efficiency and the accuracy of search results.

[0003] Currently, there are no good solutions for improving the accuracy of search results. Summary of the Invention

[0004] This application provides a search processing method, apparatus, electronic device, computer-readable storage medium, and computer program product that can accurately identify search intent and thus improve the accuracy of search results.

[0005] The technical solution of this application embodiment is implemented as follows: This application provides a search processing method, including: In response to an information input operation, the query information entered through the information input operation is displayed; The system displays multiple suggestions corresponding to the query information, wherein each suggestion is treated as a separate option. In response to a selection operation, the search results corresponding to at least one of the suggested words selected by the selection operation are displayed.

[0006] This application provides a search processing apparatus, including: An input module is used to respond to an information input operation and display the query information entered through the information input operation; The display module is used to display multiple prompt words corresponding to the query information, wherein each prompt word is treated as a separate option; The search module is configured to, in response to a selection operation, display search results corresponding to at least one of the prompt words selected by the selection operation.

[0007] This application provides an electronic device, including: Memory is used to store executable instructions for a computer; The processor, when executing computer-executable instructions stored in the memory, implements the search processing method provided in the embodiments of this application.

[0008] This application provides a computer-readable storage medium storing computer-executable instructions for inducing a processor to execute and implement the search processing method provided in this application.

[0009] This application provides a computer program product, including a computer program or computer executable instructions, which, when executed by a processor, implements the search processing method provided in this application.

[0010] The embodiments of this application have the following beneficial effects: Based on the input query information, corresponding suggestions are displayed, and different suggestions are provided as alternatives, allowing users to flexibly select one or more search terms according to their search intent. This ensures that the search results obtained with the search terms match the search intent, improving search efficiency and accuracy. Attached Figure Description

[0011] Figure 1 This is a schematic diagram illustrating the application mode of the search processing method provided in the embodiments of this application; Figure 2 This is a schematic diagram of the structure of the terminal device 400 provided in the embodiments of this application; Figures 3A to 3K This is a flowchart illustrating the search processing method provided in an embodiment of this application; Figure 4A This is a schematic diagram of the search page provided in an embodiment of this application; Figure 4B This is a schematic diagram of a word cloud provided in an embodiment of this application; Figure 4C This is a schematic diagram of the search page provided in an embodiment of this application; Figure 4D This is a schematic diagram of a word cloud provided in an embodiment of this application; Figure 4E This is a schematic diagram of the search page provided in an embodiment of this application; Figure 4F This is a schematic diagram of a word cloud provided in an embodiment of this application; Figure 5A This is a schematic diagram of a word cloud provided in an embodiment of this application; Figure 5B This is a schematic diagram of the second control provided in an embodiment of this application; Figure 5C This is a schematic diagram of the search page provided in an embodiment of this application; Figures 5D to 5EThis is a schematic diagram of the second control provided in an embodiment of this application; Figure 5F This is a schematic diagram of the search page provided in an embodiment of this application; Figures 5G to 5I This is a schematic diagram of the second control provided in an embodiment of this application; Figure 6A This is an optional flowchart illustrating the search processing method provided in an embodiment of this application; Figure 6B This is a flowchart illustrating the search processing method provided in an embodiment of this application; Figure 6C This is a schematic diagram of the search page provided in an embodiment of this application; Figures 6D to 6F This is a schematic diagram of the second control provided in an embodiment of this application. Detailed Implementation

[0012] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0013] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0014] In the following description, the terms "first, second, third" are used merely to distinguish similar objects and do not represent a specific ordering of objects. It is understood that "first, second, third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0015] It should be noted that in the embodiments of this application, user information, user feedback data and other related data are involved. When the embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0016] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0017] Before providing a further detailed description of the embodiments of this application, the nouns and terms involved in the embodiments of this application will be explained, and the nouns and terms involved in the embodiments of this application shall be interpreted as follows.

[0018] 1) Keywords: Words obtained based on query information, which can be obtained by splitting the query information and extracting features.

[0019] 2) Word cloud: A two-dimensional area composed of recommended keywords and candidate words. The two-dimensional area expands outward from the central keyword. Each recommended word in the word cloud is close to and related to each other, thus forming a graph-like effect of many keywords.

[0020] 3) Word graph: It includes multiple nodes connected by lines. Each node represents a word. The distance between nodes (the number of edges traversed from one node to another) reflects the distance between the vector representations of the words represented by the nodes. That is, the higher the relevance of the words represented by any two nodes, the smaller the distance between the two nodes.

[0021] 4) Knowledge Graph (KG): A knowledge graph is the most efficient way to represent relationships, forming a relational network by connecting all different kinds of information. A knowledge graph is a semantic network that reveals the relationships between entities, providing a formal description of things in the real world and their interrelationships. Nodes in the graph are represented by "entities," and edges by "relationships." In this embodiment, the entity corresponding to each node in the knowledge graph can be represented by a vocabulary. Based on keywords in the knowledge graph and their associated nodes, candidate words for the keywords can be determined.

[0022] This application provides a search processing method, a search processing apparatus, an electronic device, a computer-readable storage medium, and a computer program product, which can accurately identify search intent and thus improve the accuracy of search results.

[0023] The following describes exemplary applications of the electronic devices provided in the embodiments of this application. These electronic devices can be implemented as various types of user terminals, such as laptops, tablets, desktop computers, set-top boxes, mobile devices (e.g., mobile phones, portable music players, personal digital assistants, dedicated messaging devices, portable gaming devices), and in-vehicle terminals, or as servers. Exemplary applications when the device is implemented as a terminal device or a server will be described below.

[0024] refer to Figure 1 , Figure 1This is a schematic diagram illustrating the application mode of the search processing method provided in this application embodiment; for example, the servers involved include: database 500, server 200, network 300, and terminal device 400. Terminal device 400 is connected to server 200 through network 300, which can be a wide area network, a local area network, or a combination of both.

[0025] For example, server 200 could be a search engine server, users could be users of the search engine, and database 500 could be an encyclopedia database. The following explanation uses this example as a starting point.

[0026] The user enters query information in the first control 101 of the search page 103 via the terminal device 400. Based on the query information, the second control 102 displays at least one keyword related to the query information and multiple candidate words related to the keyword. In response to the selection operation of at least one word in the second control 102, the selected word is used as the search term. The terminal device 400 sends a search request carrying the search term to the server 200. The server 200 retrieves the search results corresponding to the search term from the database based on the search request and returns the search results to the terminal device 400, where the search results corresponding to the search term are displayed.

[0027] For example, server 200 could be a server for a vehicle networking platform, the user could be a user performing a place name search, and database 500 could store map data. The following explanation uses the example above as a starting point.

[0028] Without a specific travel plan, the user enters query information (e.g., "Where to go this weekend?") in the first control 101 of the search page 103 via terminal device 400. Based on the query information, the second control 102 displays at least one keyword related to the query information (e.g., the name of an entertainment venue in the same city as the user) and multiple candidate words related to the keyword (e.g., functions and activities related to the entertainment venue). If the user is interested in certain activities or locations, they can select the corresponding candidate words as search terms. In response to the selection operation of at least one word in the second control 102, the selected word is used as the search term. Terminal device 400 sends a search request carrying the search term to server 200. Server 200 retrieves the search results corresponding to the search term from database 500 based on the search request and returns the search results to terminal device 400, where the search results corresponding to the search term are displayed.

[0029] This application embodiment can be implemented using blockchain technology. The search results of this application embodiment can be used as the detection results, uploaded to the blockchain for storage, and the reliability of the detection results is ensured through a consensus algorithm. Blockchain is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include a blockchain underlying platform, a platform product service layer, and an application service layer.

[0030] This application embodiment can be implemented using database technology. A database, simply put, can be viewed as an electronic filing cabinet storing electronic files, where users can perform operations such as adding, querying, updating, and deleting data. A "database" is a collection of data stored together in a certain way, capable of being shared by multiple users, having minimal redundancy, and being independent of application programs.

[0031] A Database Management System (DBMS) is a computer software system designed to manage databases, generally possessing basic functions such as storage, retrieval, security, and backup. DBMSs can be classified according to the database model they support, such as relational or XML (Extensible Markup Language); or according to the type of computer they support, such as server clusters or mobile devices; or according to the query language used, such as Structured Query Language (SQL) or XQuery; or according to performance priorities, such as maximum scale or maximum operating speed; or other classification methods. Regardless of the classification method used, some DBMSs can cross categories, for example, simultaneously supporting multiple query languages.

[0032] In some embodiments, the server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The terminal device can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, in-vehicle terminal, etc., but is not limited to these. The terminal device and the server can be directly or indirectly connected via wired or wireless communication, which is not limited in this embodiment.

[0033] This application embodiment can also be implemented using cloud technology. Cloud technology is a general term for network technology, information technology, integration technology, management platform technology, and application technology based on cloud computing business models. It can form a resource pool, available on demand, offering flexibility and convenience. Cloud computing technology will become a crucial support. Backend services of technical network systems require substantial computing and storage resources, such as video websites, image websites, and many portal websites. With the rapid development and application of the internet industry, and driven by demands for search services, social networks, mobile commerce, and open collaboration, every item may eventually possess its own hash-coded identification mark, requiring transmission to a backend system for logical processing. Data at different levels will be processed separately, and various industry data will require robust system support, which can only be achieved through cloud computing.

[0034] See Figure 2 , Figure 2 This is a schematic diagram of the structure of the electronic device 400 provided in the embodiments of this application. The electronic device may be a terminal device or a server. Figure 2 The illustrated electronic device 400 includes at least one processor 410, a memory 450, at least one network interface 420, and a user interface 430. The various components in the electronic device 400 are coupled together via a bus system 440. It is understood that the bus system 440 is used to implement communication between these components. In addition to a data bus, the bus system 440 also includes a power bus, a control bus, and a status signal bus. However, for clarity, ... Figure 2 The general labeled all buses as Bus System 440.

[0035] Processor 410 can be an integrated circuit chip with signal processing capabilities, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or any conventional processor, etc.

[0036] User interface 430 includes one or more output devices 431 that enable the presentation of media content, including one or more speakers and / or one or more visual displays. User interface 430 also includes one or more input devices 432, including user interface components that facilitate user input, such as a keyboard, mouse, microphone, touch screen display, camera, other input buttons and controls.

[0037] The memory 450 may be removable, non-removable, or a combination thereof. Exemplary hardware devices include solid-state storage, hard disk drives, optical disk drives, etc. The memory 450 may optionally include one or more storage devices physically located away from the processor 410.

[0038] The memory 450 may include volatile memory or non-volatile memory, or both. The non-volatile memory may be read-only memory (ROM), and the volatile memory may be random access memory (RAM). The memory 450 described in this application embodiment is intended to include any suitable type of memory.

[0039] In some embodiments, memory 450 is capable of storing data to support various operations, examples of which include programs, modules, and data structures or subsets or supersets thereof, as illustrated below.

[0040] Operating system 451 includes system programs for handling various basic system services and performing hardware-related tasks, such as the framework layer, core library layer, driver layer, etc., for implementing various basic business functions and handling hardware-based tasks; The network communication module 452 is used to reach other electronic devices via one or more (wired or wireless) network interfaces 420, exemplary network interfaces 420 including: Bluetooth, WiFi, and Universal Serial Bus (USB), etc. Presentation module 453 is configured to enable the presentation of information (e.g., a user interface for operating peripheral devices and displaying content and information) via one or more output devices 431 associated with user interface 430 (e.g., a display screen, a speaker, etc.). The input processing module 454 is used to detect and translate one or more user inputs or interactions from one or more input devices 432. In some embodiments, the search processing apparatus provided in this application can be implemented in software. Figure 2A search processing device 455 stored in memory 450 is shown. This device can be software in the form of programs and plug-ins, and includes the following software modules: an input module 4551, a display module 4552, and a search module 4553. These modules are logically linked and can therefore be arbitrarily combined or further separated according to their implemented functions. The functions of each module will be described below.

[0041] The search processing method provided in this application will be described in conjunction with exemplary applications and implementations of the terminal devices provided in the embodiments of this application.

[0042] See Figure 3A , Figure 3A This is a flowchart illustrating the search processing method provided in the embodiments of this application. The following is a schematic diagram of the process. Figure 1 The terminal device 400 in the middle is the execution entity, and will combine Figure 3A The steps shown are explained.

[0043] In step 301, in response to the information input operation, the query information entered through the information input operation is displayed.

[0044] For example, query information and suggestion words can be integrated into one control, or into two different controls, namely: a first control containing a search box and a search button (the first control receives information input and displays the corresponding query information), and a second control displaying multiple suggestion words. The following explanation uses the first and second controls as examples. The first control can be a search engine control. The second control can be a smart graph. The user is a user of the search engine; this application's embodiments are explained in conjunction with the above examples.

[0045] For example, the query information can be in English, Chinese, Pinyin, etc. The search processing method provided in this application embodiment is applicable to multiple languages, such as English and Chinese.

[0046] refer to Figure 4A , Figure 4A This is a schematic diagram of a search page provided in an embodiment of this application; the search page 404A includes a first control 401A, a second control 402A, and a third control 403A. The first control 401A includes a search box 408A and a search button 409A. The search box 408A is used to input query information 407A; for example... Figure 4A The question "How to lose weight quickly without exercising?" can be answered by typing, pasting, or using voice input.

[0047] Continue to refer to Figure 3A In step 302, multiple prompts corresponding to the query information are displayed.

[0048] For example, each prompt word is displayed as a separate option, and each option can be selected individually or multiple options can be selected simultaneously (i.e., multi-select). Individual options can be displayed in the following ways: Method 1: Each tooltip is displayed in its corresponding child control as an identifier. The child control is represented by a bounding box (e.g., rectangle, ellipse). The bounding box of the selected tooltip is displayed differently from that of the unselected tooltip. For example, the bounding box (or its contents) may be a different color, the bounding box may be bolded, or the bounding box (or its contents) may be highlighted. (See reference) Figure 4A , Figure 4A This is a schematic diagram of the search page provided in the embodiment of this application; each suggestion word (i.e., keyword 405A, candidate word 406A) in the second control 402A corresponds to a sub-control, and the bounding box of the sub-control corresponding to the selected word is displayed in a different color to distinguish it from the unselected word.

[0049] Method 2: Each tooltip is displayed as a single display object. In response to a selection action (e.g., a click), the selected tooltip is shown. Selected tooltips are displayed differently from unselected tooltips, such as using different colors, larger font sizes, bolding, or highlighting. (See reference) Figure 6C , Figure 6C This is a schematic diagram of the search page provided in the embodiment of this application; each prompt word in the second control 402A is displayed as a separate display object. The prompt word 601C "don't want to exercise" is not selected and is displayed in normal form. The prompt word 602C "quick weight loss" is selected to distinguish the font display from other prompt words, and the prompt word 602C is displayed in bold.

[0050] In some embodiments, the prompts can be arranged according to any of the following attributes (arranged based on a queue or sorted based on the direction of arrangement): Attribute 1, Type. Sort by queue or arrangement direction; each queue or arrangement direction corresponds to a different type.

[0051] Attribute 2, Query Popularity. The query popularity range of the suggested terms is divided into multiple distinct intervals. Suggested terms are arranged in the queue or along the direction corresponding to the interval to which their query popularity belongs. For example: clockwise, the query popularity intervals for each arrangement direction are sorted in descending order; from left to right, the query popularity intervals for each queue are sorted in descending order, and the suggested terms within each queue are sorted according to their descending query popularity.

[0052] Attribute 3: First Letter. Based on the first letter, the prompts are divided into multiple groups, each group arranged in a different direction or queue. For example: using "abcd" as the first letter for the first direction, "efgh" as the first letter for the second direction, and so on, prompts starting with "abcd" are displayed in the first direction. Another example: from left to right, each queue corresponds to "abcd" in the first queue, "efgh" in the second queue, and so on, with prompts starting with "abcd" displayed in the first queue, and the prompts in the first queue sorted in descending order from "a" to "d".

[0053] The following explanation uses type as an example. In some embodiments, refer to Figure 6B , Figure 6B This is a flowchart illustrating the search processing method provided in this application embodiment; step 302 can be implemented through the following steps 3021A to 3022A, which are described in detail below.

[0054] In step 3021A, prompt words belonging to multiple types are displayed in multiple arrangement directions.

[0055] For example, at least one type of prompt word is distributed in each arrangement direction. When the number of multiple types n is less than or equal to the type number threshold N, n types of prompt words are distributed one-to-one in the n arrangement directions; when the number of multiple types n is greater than the type number threshold N, N-1 types of prompt words are distributed one-to-one in the first N-1 arrangement directions, and n-N+1 types of prompt words are distributed in the Nth arrangement direction.

[0056] For example: Suppose the number of types n is 6, and the type threshold N is 6. Divide the 360 ​​degrees into six 60-degree sector regions, with each sector corresponding to a sorting direction, and each sorting direction corresponding to a candidate word of a type. (Reference) Figure 6D , Figure 6D This is a schematic diagram of the second control provided in the embodiments of this application. The second control 602D displays multiple prompt words 601D. The direction pointed to by the category center line 604D is the arrangement direction. The prompt words 601D of each type are distinguished by the category boundary 603D. The prompt words of type 1 to type 6 correspond to different arrangement directions.

[0057] Assuming the quantity n is 7, which is greater than the type quantity threshold of 6, then in the 1st to 5th arrangement directions corresponding to types 1 to 5, there are prompt words corresponding to types 1 to 5 respectively. In the 6th arrangement direction, there are candidate words for types 6 and 7 respectively. (Reference) Figure 6E , Figure 6EThis is a schematic diagram of the second control provided in the embodiments of this application. In the first to fifth arrangement directions corresponding to the first to fifth types, there are prompt words corresponding to the first to fifth types respectively. The prompt word 601E corresponding to type 7 and the prompt word corresponding to type 6 are both distributed in the sixth arrangement direction.

[0058] Alternatively, in step 3022A, prompt words belonging to multiple types are displayed in multiple queues respectively.

[0059] refer to Figure 6F , Figure 6F This is a schematic diagram of the second control provided in an embodiment of this application. The queue can be represented in tabular form. The second control 602F includes multiple queues, each queue corresponding to a different delete control 603F. The delete control 603F can be used to hide the corresponding queue. Each queue corresponds to a different up-search control 604F and down-search control 606F. The up-search control 604F is used to browse the prompts in the queue upwards, and the down-search control 606F is used to browse the prompts in the queue downwards.

[0060] For example, multiple queues are in the form of multiple rows or columns, with at least one type of prompt word distributed in each queue. Multiple queues can be represented in tabular form. When the number of types n is less than or equal to the type number threshold N, n types of prompt words are distributed one-to-one in the n arrangement directions; when the number of types n is greater than the type number threshold N, N-1 types of prompt words are distributed one-to-one in the first N-1 arrangement directions, and n-N+1 types of prompt words are distributed in the Nth arrangement direction.

[0061] For example: if the threshold N for the number of types is 3, and the number of types is 3, then there will be candidate words of 3 different types distributed one-to-one in the 3 queues. Another example: (continue to refer to...) Figure 6F If the number of types is 4 (including type 1, type 2, type 3, and type 4), which is greater than the type number threshold of 3, then the first two queues of the second control 602F contain two types of prompt words (type 1: prompt word 611, prompt word 612, prompt word 613; type 2: prompt word 621, prompt word 622, prompt word 623), and the third queue contains prompt words of types 3 and 4. In the third queue, prompt words 631, 641, and 632 corresponding to types 3 and 4 are arranged alternately.

[0062] The execution order of steps 3022A and 3021A is not important.

[0063] In some embodiments, the multiple prompt words include: at least one keyword corresponding to the query information, and multiple candidate words related to the at least one keyword. The keyword is also multiple recommended words obtained based on the query information, and the keyword can be determined in the following ways: Way 1: Directly extracted from the query information, that is, the keyword comes from the query information. For example, the query information is decomposed into multiple words, and some of the words are extracted as keywords. Based on "how to lose weight quickly without exercise", "without exercise" and "lose weight quickly" can be extracted.

[0064] Way 2: Obtained by natural language semantic understanding of the query information. The keyword is not necessarily included in the query information. For example, natural language processing (NLP) is performed on the query information. Based on "how to lose weight quickly without exercise", "how to lose weight" can be extracted.

[0065] Way 3: When the query information is pinyin or letters, the keyword is the corresponding text obtained by analyzing the pinyin or letters (for example: Chinese or English, depending on the language setting of the search engine). For example, the input query information is "JF" or "jianfei", "jianf", etc. Based on the query information, the keyword corresponding to the pinyin can be obtained, "lose weight". Another example: The input query information is the letter "NLP", and based on the query information, the corresponding keywords "Natural Language Processing" and "自然语言处理" can be obtained.

[0066] Exemplarily, the candidate word is a word related to the keyword and can be obtained based on the knowledge graph. For example, the words corresponding to the nodes associated with the keyword in the knowledge graph are obtained, and the obtained words are used as candidate words for the keyword.

[0067] In some embodiments, the keyword and candidate word displayed in the second control can be updated in real time or displayed when the information input operation ends. Real-time update means that it is updated in real time according to the characters input in the first control. Whenever a character is input, a character is displayed in the first control, and the terminal device recognizes the input character as query information and displays the keyword and candidate word in the second control. Display when the information input operation ends (for example, when an input operation is detected and subsequent input operations pause), that is, according to the characters that have been input, the corresponding keyword and candidate word are displayed.

[0068] In some embodiments, the candidate words can be arranged according to any of the following different attributes (arranged based on a queue or sorted based on the arrangement direction): Attribute 1: Type. Sort through a queue or arrangement direction, and each queue or arrangement direction corresponds to a different type.

[0069] Attribute 2, Query Popularity. The query popularity range of candidate terms is divided into multiple distinct intervals. Candidate terms are arranged in queues or along the direction corresponding to their respective query popularity intervals. For example: clockwise, the query popularity intervals for each arrangement direction are sorted in descending order; from left to right, the query popularity intervals for each queue are sorted in descending order, and the candidate terms in each queue are sorted according to their descending query popularity.

[0070] Attribute 3: First Letter. Candidate words are divided into multiple groups based on their first letter, with each group arranged in a different direction or queue. For example, "abcd" is used as the first letter for the first direction, "efgh" for the second direction, and so on. Candidate words whose first letter is "abcd" are displayed in the first direction. Another example: from left to right, each queue corresponds to "abcd" in the first queue, "efgh" in the second queue, and so on. Candidate words whose first letter is "abcd" are displayed in the first queue, and the candidate words in the first queue are sorted in descending order from "a" to "d".

[0071] In some embodiments, reference Figure 3B , Figure 3B This is a flowchart illustrating the search processing method provided in this application embodiment; step 3021A can be implemented through steps 3021B to 3022B, as detailed below.

[0072] In step 3021B, at least one keyword is displayed in the second control.

[0073] For example, at least one keyword is located in the center of the second control. See also... Figure 4A Keyword 405A (e.g., "quick weight loss", "don't want to exercise") is a different color from candidate word 406A and is displayed in the center of the second control 402A. The method for obtaining keywords has been explained above and will not be repeated here.

[0074] In step 3022B, candidate words belonging to multiple types are displayed in multiple arrangement directions outward from at least one keyword.

[0075] For example, at least one type of candidate word is distributed in each arrangement direction, and each type of candidate word is associated with at least one keyword. Each arrangement direction corresponds to an angle range in the second control with at least one keyword as its endpoint, and the rays with the maximum and minimum angles are the two boundary lines of the arrangement direction. (Reference) Figure 5A , Figure 5AThis is a schematic diagram of a word cloud provided in an embodiment of this application; the word cloud is a region composed of multiple keywords and candidate words, where keyword 501A (keyword 501A refers to at least one keyword) is an example of a word cloud. Figure 5A (This is for illustrative purposes only and does not impose a limit on the number of keywords.) Located at the center, i.e., the endpoint of the arrangement direction, multiple candidate words 505A are arranged outwards from the center of keyword 501A. Different types of candidate words 505A correspond to different category sectors 503A. Each category sector 503A includes a category midline 504A, and different category sectors 503A are separated by category boundaries 502A. The arrangement direction is the direction of the arrow corresponding to the category midline 504A. The two boundary lines of category sector 503A are the rays corresponding to the maximum and minimum angles, respectively.

[0076] For example, candidate words of at least one type are distributed in each arrangement direction, including: when the number of multiple types n is less than or equal to the type number threshold N, there are n candidate words of n types distributed one-to-one in the n arrangement directions; when the number of multiple types n is greater than the type number threshold N, there are N-1 candidate words of N-1 types distributed one-to-one in the first N-1 arrangement directions, and n-N+1 candidate words of n types are distributed in the Nth arrangement direction.

[0077] For example, suppose the number of types n is 4, and the type number threshold N is 4. Then, in the four arrangement directions (e.g., up, down, left, right) with the keyword as the endpoint, candidate words corresponding to the four types are distributed one by one. The rays (classification midlines) corresponding to the four arrangement directions divide the 360 ​​degrees into four 90-degree regions with the keyword as the center. Suppose the number n is 6, which is greater than the type number threshold 4. Then, in the first to third arrangement directions corresponding to the first to third types, there are candidate words corresponding to the first to third types, respectively. In the fourth arrangement direction, there are candidate words for the fourth, fifth, and sixth types.

[0078] In some embodiments, candidate words and keywords may be randomly distributed in the second control, for example, sorted according to the order of the initials of the first character's pinyin, or sorted according to the number of strokes.

[0079] In some embodiments, the shape of the area used to display keywords in the second control can be rectangular, circular, sector-shaped, elliptical, or elliptical-sector-shaped, etc., and the area used to display keywords in the second control can be geometrically arranged in different directions based on a center. (Reference) Figure 5C , Figure 5C This is a schematic diagram of the search page provided in the embodiment of this application; the second control 502C in the search page 404A is elliptical, and the keyword 501C (keyword 501C refers to at least one keyword) is used. Figure 5A(This is for illustrative purposes only and does not limit the number of keywords.) Arrange candidate words in different directions from the center (503C) (505C).

[0080] In some embodiments, if the region is a sector or an elliptical sector, the keyword is located at the central corner of the sector, and candidate words are arranged in an arc from the keyword outwards within the sector. If the region is a rectangle, a circle, or the center of a circle, the keyword is located at the center of the rectangle or circle, and candidate words are arranged outwards from the keyword.

[0081] In some embodiments, in the second control, the following two parameters for each type of candidate word are positively correlated: the distance between the candidate word and at least one keyword, and the relevance of the candidate word to at least one keyword.

[0082] For example, the features of at least one keyword should be considered as a whole, such as using the mean of vector representations; relevance, which is the reciprocal of the distance represented by vectors (e.g., Euclidean or cosine distance). For example, obtain the first word vector corresponding to at least one keyword, and the second word vector for each candidate word. Calculate the cosine distance between each second word vector and the first word vector. Use the cosine distance to represent the relevance between the candidate word and the keyword. The candidate word with the closest cosine distance is located closest to the keyword. The farther away from the keyword, the greater the cosine distance between the candidate word and the keyword.

[0083] In some embodiments, the distance between a candidate word and at least one keyword is positively correlated with at least one of the following parameters of the candidate word: 1. Search frequency; for example, the higher the frequency, the closer it is to the keyword; 2. The order of the first character (pinyin / letter); for example, the earlier the letter corresponding to the first letter is in the alphabetical order, the closer it is to the keyword. Alternatively, the smaller the number corresponding to the first letter, the closer it is to the keyword.

[0084] In some embodiments, at least one of the following display methods is applied to distinguish different types of candidate words: Method 1: Apply different display parameters to different types of candidate words; for example, use different display methods (color, font size, font) for words belonging to different types at boundaries in different directions. For instance, at the boundary between Type 1 and Type 2, Type 1 words are displayed in blue, and Type 2 words are displayed in green; different display methods are applied to different types of words as a whole. (Reference) Figure 5D , Figure 5D This is a schematic diagram of the second control provided in the embodiments of this application; the second control 502D includes a control surrounding the keyword 501D (keyword 501D refers to at least one keyword). Figure 5D (This is for illustrative purposes only and does not limit the number of keywords.) Multiple different candidate words 505D are displayed, with each type of candidate word showing a different color or pattern.

[0085] Method 2: Display the dividing line between any two adjacent arrangement directions. (See also...) Figure 5D Each type of candidate word corresponds to a different region, and there is a dividing line 503D between each region.

[0086] In some embodiments, within the display area (the shape of the display area includes rectangles, circles, sectors, ellipses, elliptical sectors, and triangles) formed around at least one keyword in the second control, the arrangement direction of each type satisfies one of the following conditions: Condition 1: The span of the angle interval corresponding to the arrangement direction of each type is the same, and the span is the ratio of the following two parameters: the span of the angle interval of the display area, and the number of candidate word types related to at least one keyword. That is, the types have the same weight.

[0087] For example, if the display area is a circle corresponding to a 360-degree angle range, and the display area includes 4 types of candidate words, then the span of the angle range corresponding to the arrangement direction of each type is 90 degrees, which is the ratio between 360 degrees and the number of candidate word types, 4.

[0088] Condition 2: The span of the angular interval corresponding to the arrangement direction of each type is the product of the following two parameters: the span of the angular interval of the display area, and the proportion of the number of candidate words of the type in all types. That is, the weight of a type is positively correlated with the number of candidate words included in the type.

[0089] For example, suppose there are three types: Type 1, Type 2, and Type 3. The percentage of candidate words for each type in the total number of candidate words is as follows: Type 1, 50%; Type 2, 25%; and Type 3, 25%. If the angular range of the display area is 180 degrees, then the angular range corresponding to the arrangement direction of Type 1 is 90 degrees, the angular range corresponding to the arrangement direction of Type 2 is 45 degrees, and the angular range corresponding to the arrangement direction of Type 3 is 45 degrees.

[0090] In some embodiments, reference Figure 3C , Figure 3C This is a flowchart illustrating the search processing method provided in this application embodiment; during step 3022B, the displayed candidate words can also be deleted through the following steps 3023B to 3025B, as detailed below.

[0091] In step 3023B, a delete button is displayed in each arrangement direction.

[0092] refer to Figure 5E , Figure 5EThis is a schematic diagram of the second control provided in the embodiment of this application; in the second control 502D, a deletion control 501E is displayed in the sorting direction of the candidate words corresponding to each type.

[0093] In step 3024B, in response to the triggering operation of the first button in the first arrangement direction, for the candidate word of the first type corresponding to the first arrangement direction, one of the following processes is performed: the candidate word of the first type is replaced with the candidate word of the second type.

[0094] For example, the second type differs from the multiple types in that its relevance to keywords is second only to the multiple types. For instance: each arrangement direction is closer to the center or edge (see reference). Figure 5E A delete button can be set at the position of the delete control 501E (located at the edge of the second control 502D). If triggered, the current arrangement direction is deleted and replaced by other types (not previously displayed), thus allowing more words of other types to be displayed; if there are other types that are not currently displayed, they are replaced with other types.

[0095] Assumption: Reference Figure 5E Based on a knowledge graph, seven candidate word categories are calculated for each keyword. Six categories are displayed, while the seventh category is hidden. The angle range for each category's arrangement direction spans 60 degrees. When a category of words is deleted, the seventh category of words is displayed.

[0096] In step 3025B, the display of candidate words of the first type is stopped, and the angle ranges of other arrangement directions are updated to cover the first arrangement direction.

[0097] For example, other arrangement directions are those other than the first arrangement direction among multiple arrangement directions. For instance, a delete button can be set near the center / edge of each arrangement direction. If triggered, the arrangement direction is deleted, and the remaining types are reassigned arrangement directions (the angle range becomes larger).

[0098] Assumptions: Based on the graph, 6 categories are calculated, and 6 categories of words can be displayed. The angular range of the arrangement direction corresponding to each category is 60 degrees. When a category of words is deleted, the remaining 5 categories of words have an angular range of 72 degrees for each category's arrangement direction.

[0099] In some embodiments, reference Figure 3D , Figure 3D This is a flowchart illustrating the search processing method provided in this application embodiment; during step 3022B, candidate words can also be updated through steps 3026B to 3027B, as detailed below.

[0100] In step 3026B, in response to a browsing operation along the first arrangement direction in the second control, the content distributed in the first arrangement direction is moved along the first arrangement direction.

[0101] For example, the first arrangement direction can be any direction. The content distributed in the first arrangement direction can be at least one keyword and multiple candidate words, and at least one keyword will gradually disappear as you move along the first arrangement direction.

[0102] refer to Figure 4B , Figure 4B This is a schematic diagram of a word cloud provided in an embodiment of this application; assuming that keyword 402B and candidate word 403B can form a word cloud 401B, and the second control 402A displays the part of word cloud 401B centered on keyword 402B, and in response to the user's sliding operation (browsing operation) on the second control, the other parts of candidate word 403B are displayed.

[0103] In step 3027B, new candidate words of the first type are displayed in the empty area formed by moving at least part of the content out in the first arrangement direction.

[0104] For example, the distance moved is positively correlated with the magnitude of the browsing operation, the first arrangement direction is the arrangement direction corresponding to the browsing operation, and the first type is the type corresponding to the first arrangement direction.

[0105] For example, browsing operations can take the form of swiping or dragging; the number of directional controls used to receive browsing operations can be a multiple of 4, such as controls for the four directions: up, down, left, and right. These four directions can be further divided evenly into multiple directional controls.

[0106] In some embodiments, reference Figure 3G , Figure 3G This is a flowchart illustrating the search processing method provided in this application embodiment; new candidate words of the first type are determined through the following steps 3041G to 3044G, which are described in detail below.

[0107] In step 3041G, the word graph is queried based on the edge keyword and at least one keyword, and all nodes of the next level corresponding to the edge keyword and at least one keyword are found in the word graph.

[0108] For example, edge keywords are candidate words located at the edges of the first sorting direction before responding to a browsing action; a word graph, also known as a knowledge graph, has each node corresponding to an entity word, and the edges between nodes represent the relationships between them. The next-level node is the node that has only one edge to the current node.

[0109] In step 3042G, the co-occurrence counts of all nodes at the next level of the edge keyword and at least one keyword are determined, and the nodes are sorted in descending order according to the co-occurrence counts to obtain the descending order sorting result.

[0110] For example, co-occurrence means appearing together, and the co-occurrence count refers to the number of times a keyword appears repeatedly. For instance, we have keywords A, B, and C, where keyword A is a marginal keyword. Keyword A has 13 next-level nodes, keyword B has 15, and keyword C has 10. The total number of nodes is 38. Node 1 appears 3 times, so node 1 has a co-occurrence count of 3. This process continues, obtaining the co-occurrence count for each node, and then sorting each node in descending order based on the co-occurrence count to obtain the sorted result.

[0111] In step 3043G, among the multiple nodes at the head of the descending sort result, nodes that are semantically opposite to the edge keywords and at least one keyword are deleted, and candidate words of the first type that have already been displayed are also deleted.

[0112] For example, in the descending sort results, among the multiple nodes at the top, the semantics of the candidate words for each node are analyzed to identify nodes with semantics opposite to the keywords, and the corresponding nodes are deleted. The candidate words corresponding to each node are compared with the already displayed candidate words; if the comparison results are the same, the candidate words corresponding to the node have been displayed, and the node is deleted.

[0113] For example, if the candidate word corresponding to a node is already displayed in the second control, then the node is deleted; if the candidate word corresponding to a node has a semantic opposite to the edge keyword and at least one keyword, then the node is also deleted. For example: if the keywords are "don't want to exercise" and "quick weight loss", and the knowledge graph of the keyword "quick weight loss" in the weight loss field yields nodes related to exercise such as "fitness" and "Pilates", but these nodes have a semantic opposite to the keyword "don't want to exercise", then the nodes corresponding to the candidate words "fitness" and "Pilates" are deleted.

[0114] In step 3044G, the word represented by at least one of the remaining nodes is taken as a new candidate word of the first type.

[0115] In some embodiments, if there are no remaining nodes after step 3043G, the next-level nodes of the candidate words in the word graph can be queried based on the candidate words at the edge positions, the co-occurrence frequency can be counted and sorted, and the antonyms of the keywords and words that have already been displayed in the sliding arrangement direction can be filtered out to obtain new candidate words.

[0116] In some embodiments, if the browsing operation is not the first browsing operation received, and after step 302, in response to a browsing operation that crosses from the first arrangement direction to the second arrangement direction in the second control, candidate words of the type corresponding to the second arrangement direction are displayed in the area between the current browsing position and at least one keyword.

[0117] For example, when a user performs multiple vertical swipes, after the terminal device updates the display vertically, if the user moves horizontally and there are many gaps between the current browsing position and the keywords, multiple candidate words are displayed layer by layer in these gaps, from near to far. (Reference) Figure 5B , Figure 5B This is a schematic diagram of the second control provided in this application embodiment; the solid line of the second control 502B represents the initial position of the second control 502B in the word cloud, and the arrows represent the direction of movement of each browsing operation (browsing operation 503B, browsing operation 504B, browsing operation 505B). That is, two browsing operations were first performed on the first direction 507B (corresponding to the first arrangement direction), and then one browsing operation was performed on the horizontal direction, crossing to the second direction 508B (corresponding to the second arrangement direction). Region 506B is the area between the original position of the second control 502B and the position corresponding to the second control 502B after three browsing operations. In region 506B, candidate words are filled in layer by layer from the closest to the keyword 501B to the farthest. Region 506B is the area between the current browsing position (the browsing position corresponding to browsing operation 505B) and at least one keyword.

[0118] In some embodiments, reference Figure 3E , Figure 3E This is a flowchart illustrating the search processing method provided in this application embodiment; step 3022A can be implemented through the following steps 3021E to 3022E, which are described in detail below.

[0119] In step 3021E, at least one keyword is displayed in the second control.

[0120] For example, the execution of step 3021E can be referred to 3021B above, and will not be repeated here.

[0121] In step 3022E, candidate words belonging to multiple types are displayed in multiple queues respectively.

[0122] For example, multiple queues are presented in a multi-row or multi-column format, with each queue containing candidate words of at least one type. Each candidate word type is associated with at least one keyword. Alternatively, multiple candidate words can be displayed in a table, with each row or column corresponding to a type. Keywords can be displayed in the header, center row, or column of the table. (See reference) Figure 5F , Figure 5FThis is a schematic diagram of the search page provided in an embodiment of this application; the second control 502F in the search page 404A is presented in table format. Figure 5F The second control, 502F, is displayed as a thumbnail. (Reference) Figure 5G , Figure 5G This is a schematic diagram of the search page provided in an embodiment of this application; Figure 5G This is a magnified view of the second control 502F. Each column in the table corresponds to a candidate word type, and each type has a corresponding delete control 503F, as well as controls for moving the content in each column up and down: a look-up control 504F and a look-down control 505F. Keyword 501F (Keyword 501F refers to at least one keyword). Figure 5A (This is for illustrative purposes only and does not impose a limit on the number of keywords.) It corresponds to three types: Type 1 (including candidate words 511, 512, 513, etc.); Type 2 (including candidate words 521, 522, 523, etc.); and Type 3 (including candidate words 531, 532, 533, etc.). Reference Figure 5I , Figure 5I This is a schematic diagram of the search page provided in an embodiment of this application. Keyword 501F is displayed in the middle row of the table, at the center of the second control 502F.

[0123] For example, having at least one type of candidate word distributed in each queue can be achieved as follows: when the number of multiple types n is less than or equal to the type number threshold N, n types of candidate words are distributed one-to-one in the n queues; when the number of multiple types n is greater than the type number threshold N, N-1 types of candidate words are distributed one-to-one in the first N-1 queues, and n-N+1 types of candidate words are distributed in the Nth queue.

[0124] For example, if the threshold N for the number of types is 3, and the number of types is 3, then there will be candidate words of 3 different types distributed one-to-one in the 3 queues. Another example: [reference needed] Figure 5H , Figure 5H This is a schematic diagram of the second control provided in the embodiments of this application. The number of types is 4 (including type 1, type 2, type 3, and type 4), which is greater than the threshold for the number of types. Therefore, the first two queues have candidate words of two types distributed one-to-one, and the third queue has candidate words of the third and fourth types distributed. In the third queue, candidate words 531, 541, and 532 corresponding to type 3 and type 4 are arranged alternately.

[0125] In some embodiments, reference Figure 3F , Figure 3FThis is a flowchart illustrating the search processing method provided in the embodiments of this application. In the process of implementing step 3022A based on steps 3021E to 3022E, candidate words can also be updated through steps 3023E to 3024E, as detailed below.

[0126] In step 3023E, in response to a browsing operation along the target direction in the target queue of the second control, the content in the target queue is moved along the target direction.

[0127] For example, the content in the target queue can be at least one keyword and multiple candidate words. As you move along the target direction, at least one keyword will gradually disappear. The target direction is the direction corresponding to the browsing operation. For example, the target direction could be... Figure 5F The lower direction corresponding to the 505F control moves the candidate words in the corresponding queue upwards in sequence, with new candidate words appearing from below and the displayed candidate words being hidden from above.

[0128] In step 3024E, new candidate words of the target type are displayed in the free area formed by removing at least part of the content from the target queue.

[0129] For example, the distance moved is positively correlated with the magnitude of the browsing operation, and the target type corresponds to the type of the target queue. The target arrangement direction corresponds to the arrangement direction of the browsing operation. Browsing operations include: swiping; pressing the up and down search controls in the column.

[0130] In some embodiments, reference Figure 3H , Figure 3H This is a flowchart illustrating the search processing method provided in this application embodiment; during step 3022E, the displayed candidate words can also be deleted through the following steps 3026E to 3028E, as detailed below.

[0131] In step 3026E, a delete button is displayed on each queue.

[0132] Continue to refer to Figure 5F Each queue has a delete control (503F) displayed above it. The delete control (503F) is used to delete the corresponding queue.

[0133] In step 3027E, in response to a trigger operation on a first button in the first queue, for a candidate word of the first type corresponding to the first queue, one of the following processes is performed: the candidate word of the first type is replaced with a candidate word of the second type.

[0134] For example, the second type differs from the multiple types in that its relevance to keywords is second only to the multiple types. For instance, if there are four candidate word types, and three are currently displayed, in response to a trigger action of the delete control 503F corresponding to type 3, the queue of candidate words corresponding to type 3 is hidden, and the queue of candidate words corresponding to type 4 is displayed.

[0135] In step 3028E, the display of candidate words of the first type is stopped, and the size of other queues is increased to cover the first queue.

[0136] For example, increasing the size of other queues includes increasing the width or height of the queue. Other queues are queues other than the first queue. For example: there are 3 candidate word types, and 3 are currently displayed. In response to the triggering operation of the delete control 503F corresponding to type 3, the queue of candidate words corresponding to type 3 is hidden, and the width of the remaining queues of type 1 and type 2 is adjusted to cover the position of the hidden queue corresponding to type 3.

[0137] In some embodiments, when step 302 is performed, a search intent is displayed in a second control, wherein the search intent represents the intent to search by entering query information.

[0138] For example, obtaining search intent based on query information can be achieved in the following ways: Method 1: Dictionary and Template-Based Rule-Based Approach: Different search intents will have different domain dictionaries, such as product names, location names, etc. Based on the descending order of similarity between the query information and different domain dictionaries, the domain corresponding to the top-ranked domain dictionary in the descending order is obtained, and the information corresponding to that domain is displayed as the search intent.

[0139] Method 2: Obtain the search records corresponding to the search engine based on the query information, statistically analyze the information with the highest search frequency among the information related to the query information based on the search records, and take the information with the highest search frequency as the search intent.

[0140] Method 3: Based on the query information, call the convolutional neural network model to make predictions and obtain the search intent.

[0141] In some embodiments, reference Figure 3J , Figure 3J This is a flowchart illustrating the search processing method provided in this application embodiment; multiple types of candidate words are determined through the following steps 3041J to 3044J, which are described in detail below.

[0142] In step 3041J, at least one keyword is extracted from the query information.

[0143] As an example, step 302 above has already explained how to extract keywords from the query information, so it will not be repeated here.

[0144] In step 3042J, when there are multiple keywords, all nodes at the next level of each keyword are searched from the word graph.

[0145] For example, a word graph includes a keyword knowledge graph. The next-level nodes of a keyword are nodes in the knowledge graph that have only one edge connecting them to the node corresponding to the keyword. This allows you to retrieve each next-level node corresponding to each keyword.

[0146] In step 3043J, the co-occurrence count of all nodes at the next level of multiple keywords is determined, and the nodes are sorted in descending order according to the co-occurrence count to obtain the descending sort result.

[0147] For example, the method of counting co-occurrences and sorting in descending order in step 3043J is the same as in step 3042G, and will not be repeated here.

[0148] In step 3044J, among the multiple nodes at the head of the descending sort result, the node whose semantics are opposite to any keyword is deleted, and the words represented by the remaining multiple nodes are used as candidate words of multiple types.

[0149] For example, the method of deleting nodes that have the opposite meaning to any keyword has been explained in step 3043G, and will not be repeated here.

[0150] In some embodiments, reference Figure 3K , Figure 3K This is a flowchart illustrating the search processing method provided in this application embodiment; multiple types of candidate words are determined through the following steps 3041K to 3043K, which are described in detail below.

[0151] In step 3041K, at least one keyword is extracted from the query information.

[0152] The execution of step 3041K is the same as that of step 3041J, and will not be repeated here.

[0153] In step 3042K, when the number of at least one keyword is one, all nodes of the next level of the keyword are searched from the word graph.

[0154] The method for obtaining all nodes of the next level in step 3042K is the same as that in step 3042J, and will not be repeated here.

[0155] In step 3043K, from all nodes at the next level of the keyword, nodes that are semantically opposite to any keyword are deleted, and the words represented by the remaining nodes are used as candidate words of multiple types.

[0156] For example, the method of deleting nodes that have the opposite meaning to any keyword has been explained in step 3043G, and will not be repeated here.

[0157] In this embodiment of the application, by filtering candidate words, the display of candidate words that do not match the user's intent is avoided, thereby improving the efficiency and accuracy of obtaining search results.

[0158] Continue to refer to Figure 3A In step 303, in response to the selection operation, the search results corresponding to at least one prompt word selected by the selection operation are displayed.

[0159] For example, at least one suggestion word in the selected state is used as the search term. When the suggestion word includes at least one keyword and multiple candidate words, the search term is at least one of the keyword in the selected state and multiple candidate words. The search results can be displayed in a second control. Alternatively, a third control specifically designed for displaying search results can be used; see further. Figure 4A The third control, 403A, is used to display search results.

[0160] This application embodiment determines and displays multiple keywords and candidate words associated with the keywords based on query information. Both keywords and candidate words are available for users to choose from, making the selection of search terms more flexible. When users are not familiar with the relevant field of the query information, providing users with relevant keywords and candidate words makes it easier for them to flexibly choose search terms, thereby improving the accuracy of the search.

[0161] In some embodiments, at least one keyword is selected by default, and multiple candidate words are unselected by default. Each selection operation is used to trigger the display of the corresponding search results. Step 303 can be implemented in the following way: in response to the selection operation for at least one first target word, display that at least one first target word is selected, wherein at least one first target word is at least one candidate word selected by the selection operation among multiple candidate words, and display the search results corresponding to the at least one keyword and the at least one first target word.

[0162] For example, in response to each selection action in the second control, the search results for the currently selected search term are automatically displayed in the third control.

[0163] Example, reference Figure 4C as well as Figure 4E , Figure 4C , Figure 4E This is a schematic diagram of the search page provided in an embodiment of this application; Figure 4C The candidate keyword "what to eat" is displayed as unselected. Figure 4EThe phrase "what to eat" was selected as the primary target word and used as a 401E keyword. Figure 4E The candidate words in the second control 402E are also updated to the candidate words corresponding to the keywords "what to eat", "don't want to exercise", and "quick weight loss". At the same time, the corresponding search results are displayed when "what to eat", "don't want to exercise", and "quick weight loss" are used as search terms.

[0164] In some embodiments, at least one keyword and multiple candidate words are in an unselected state by default; step 303 can be implemented as follows: in response to a selection operation for at least one third target word, display that at least one third target word is selected, wherein at least one third target word is at least one of the multiple candidate words and at least one keyword selected by the selection operation. (In response to a trigger operation for the search button in the first control, this can be omitted, i.e., the subsequent display is performed automatically.) Display search results corresponding to at least one search term, wherein at least one search term includes the first target word.

[0165] For example, when a selection action is received for a search term in the second control, the search results will not be displayed automatically; they will only be displayed after the search button is triggered.

[0166] In some embodiments, reference Figure 3I , Figure 3I This is a flowchart illustrating the search processing method provided in the embodiments of this application; after step 303, the following steps 3051 to 3052 are performed, which will be described in detail below.

[0167] In step 3051, in response to a cancellation operation for at least one second target word, it is displayed that at least one second target word is in an unselected state.

[0168] For example, at least one second target word is at least one of at least one keyword and at least one first target word that has been canceled by the cancellation operation. The unselected state can also be represented by hiding the second target word, i.e., no longer displaying the second target word.

[0169] Example, reference Figure 4E Suppose that at least one of the selected words "don't want to exercise", "lose weight quickly", and "what to eat" is selected by the cancellation operation, and the word cancelled by the cancellation operation is the second target word.

[0170] Continue to refer to Figure 3I In step 3052, the updated search results are displayed.

[0171] For example, the updated search results correspond to at least one keyword and at least one first target word, excluding at least one second target word. For instance: the selected words are Keyword 1, Candidate 2, and Candidate 3; Candidate 3 is selected as the second target word; the deselection operation removes Candidate 3; and Candidate 2 and Keyword 1 are selected as search terms. The updated search results will then be displayed based on Candidate 2 and Keyword 1. Another example: Figure 4E If you deselect "What to eat" in the menu, then... Figure 4E The content displayed in the second control 402E is restored to... Figure 4C The keywords and candidate words are displayed in the second control 402C.

[0172] In some embodiments, step 3052 is implemented by: in response to a trigger operation on the search button in the first control, displaying search results corresponding to at least one suggestion word selected by the selection operation. That is, referring to... Figure 4A By triggering the search button 409A in the first control 401A, the third control 403A displays the search results corresponding to at least one prompt word.

[0173] In some embodiments, before a selection operation is performed in the second control, search results corresponding to at least one keyword and multiple candidate words are displayed in the third control. After a subsequent selection operation is received, the previous search results are refreshed to reflect the search results corresponding to the search term. That is, when a user enters query information, the second control displays the keywords and candidate words corresponding to the query information, while the third control synchronously displays the search results corresponding to the keywords and candidate words until a selection operation is received, at which point the previous search results are refreshed to reflect the search results corresponding to the search term.

[0174] This embodiment of the application improves the real-time nature of the search by displaying corresponding search results as keywords and candidate words change, making it easier for users to determine the search content they need, improving search efficiency, and saving the operation steps required for the search. Based on the query information entered in the first control, the corresponding keywords and candidate words are displayed. Different keywords and candidate words are provided as options based on the query information, allowing users to flexibly select one or more search terms according to their search intent. This ensures that the search results obtained with the search terms match the search intent, making it easier for users to search for the information they need, saving the operation steps of users in the process of searching for information, and improving search efficiency and the accuracy of search results.

[0175] The following will describe an exemplary application of the embodiments of this application in a real-world application scenario.

[0176] When a user searches for information in an unfamiliar field, due to unfamiliarity with related terminology, they enter their query information into the search engine control (the first control mentioned above) on the search page of their terminal device. Based on this query, the intelligent graph (the second control mentioned above) displays keywords related to the query information, as well as candidate words related to those keywords. The user can select words similar to their query intent from the candidate words and keywords as search terms. In response to the selection of any word in the second control, the selected word is used as the search term, and the corresponding search results are displayed.

[0177] See Figure 6A , Figure 6A This is a flowchart illustrating the search processing method provided in the embodiments of this application. The following is a schematic diagram of the process. Figure 1 The terminal device 400 in the middle is the execution entity, and will combine Figure 6A The steps shown are explained.

[0178] In step 601A, in response to the input operation of query information, the keywords corresponding to the query information and the candidate words related to the keywords are displayed in the intelligent graph.

[0179] For example, when a user searches, the user's query information is broken down into multiple keywords, and a batch of candidate words is calculated based on these keywords and displayed in a word cloud centered on the keywords. Keyword extraction based on query information can be achieved by using natural language processing technology to break down the query information into multiple independent words, discarding unnecessary adverbs and meaningless parts of the words, and using the remaining words as keywords.

[0180] For example, based on keywords, candidate words related to those keywords can be obtained. This can be achieved by: determining the first-level node corresponding to each keyword in the knowledge graph; after obtaining the first-level nodes associated with each keyword, counting the co-occurring nodes; sorting the nodes in descending order of their frequency of occurrence; and selecting the node at the top of the descending sort. For instance, assuming the keywords "weight loss" and "lack of exercise" both contain the node "diet," the node "diet" has a higher priority and will be ranked higher.

[0181] For example, the candidate words are categorized by their direction relative to the word cloud center (whether they are within the same category sector or at the boundary between two categories), and their distance relative to the word cloud center (calculated by layer and circle). Candidate words are scored according to their priority (priority scoring can be determined based on the relevance between the candidate word and the keyword; the higher the relevance, the higher the score, and the closer the candidate word is to the keyword, e.g., placed in the inner circle), and arranged in order from closest to furthest. Reference Figure 5A , Figure 5AThis is a schematic diagram of a word cloud provided in an embodiment of this application. Keyword 501A is located in the center of the word cloud, and multiple candidate words 505A are arranged outwards from the center of keyword 501A. Different types of candidate words 505A correspond to different classification sectors 503A. Each classification sector 503A includes a classification midline 504A, and different classification sectors 503A are separated by classification boundaries 502A. The distance between multiple candidate words 505A and keyword 501A is positively correlated with the score corresponding to each candidate word 505A; that is, the higher the relevance to keyword 501A, the closer candidate word 505A is to keyword 501A in the word cloud.

[0182] refer to Figure 4A , Figure 4A This is a schematic diagram of the search page provided in an embodiment of this application; the search page 404A includes a first control 401A, a second control 402A, and a third control 403A. In the first control 401A, in response to the user's input operation on the first control 401A, the query information 407A "how to lose weight quickly without exercising" is displayed. The query information is decomposed through semantic analysis to obtain the keywords "lose weight quickly" and "don't want to exercise". The keyword 405A is displayed in the second control 402A (i.e., the intelligent graph) and highlighted to indicate that it has been automatically selected. When the keyword is clicked, it switches to an unselected state. Based on the keyword, multiple candidate words related to the keyword can be obtained, and the candidate word 406A is also displayed in the second control 402A.

[0183] In step 602A, a browsing operation for words in the smart graph is received.

[0184] Example, reference Figure 4B , Figure 4B This is a schematic diagram of a word cloud provided in an embodiment of this application. Candidate words 403B and keywords 402B are used as recommended words, forming a word cloud 401B. The recommended words (keywords and candidate words) are interconnected; the closer a recommended word is to the center of the word cloud, the stronger its association with the query information, while the further outwards, the more dispersed and in-depth it becomes. The radiating directions from the center of the word cloud represent different types of recommended keywords (recommended words). After seeing recommended words of interest in different directions, users often continue browsing along those directions, indicating a need for in-depth exploration of that recommendation type. When browsing to the edge of a direction, the position of the second control 402A within the word cloud 401B can be switched using a sliding operation to view more recommended words. Browsing operations can include sliding, dragging, etc.

[0185] In step 603A, the displayed candidate words are updated based on the direction corresponding to the browsing operation.

[0186] For example, users can swipe to browse the word cloud. As the user swipes in a direction within the second control, more and deeper recommendations in that direction are presented. When the user swipes to the edge of the word cloud, the word cloud expands its content in that direction. All content is a product of associating existing recommended words in the current direction, calculated in real time. The green recommended words in the diagram represent new words recommended in real time. Imagine that while browsing from any position, a user might become interested in a particular direction and explore further in that direction, continuously triggering new recommended words calculated in real time. All recommended content is based on the user's current interests.

[0187] refer to Figure 4C , Figure 4C This is a schematic diagram of the search page provided in an embodiment of this application; compared to Figure 4A The content in the second control 402A and the second control 402C moves to the right, see reference. Figure 4D , Figure 4D This is a schematic diagram of a word cloud provided in an embodiment of this application; in word cloud 401B, in response to a browsing operation (sliding operation) on second control 402C, second control 402C moves to the left relative to second control 402A, and then the content therein moves to the right, presenting the new candidate word 401D in word cloud 401B to the user.

[0188] For example, when a user slides to the edge of the word cloud, a new batch of words is recommended to the surrounding blank spaces from the current position. The core keyword is still the center, and the blank spaces are filled in order from near to far. Using the keyword and the candidate words at the edge position, the next level node in the word graph is queried again to count the co-occurrence frequency and sort them. Antonyms of the keyword and words that have already been displayed in the sliding direction are filtered out. The occupied positions will not be filled with other words, and the newly recommended words do not overlap with other recommended words in the word cloud.

[0189] In some embodiments, the user performs multiple swipes in the depth direction. If a swipe is performed laterally, the gaps between the new position and the center of the word cloud are filled layer by layer from near to far.

[0190] refer to Figure 5B , Figure 5BThis is a schematic diagram of the second control provided in this application embodiment; the solid line of the second control 502B represents the initial position of the second control 502B in the word cloud, and the arrows represent the direction of movement for each browsing operation (browsing operation 503B, browsing operation 504B, browsing operation 505B), that is, two browsing operations were performed in the same direction first, and then one browsing operation was performed horizontally. Region 506B is the area between the original position of the second control 502B and the corresponding position of the second control 502B after the three browsing operations. In region 506B, the corresponding candidate words are filled in layer by layer from the closest to the keyword 501B to the farthest.

[0191] During the execution of step 603A, either step 604A or step 605A may be executed. The execution order of step 604A or step 605A is not important.

[0192] In step 604A, in response to the selection operation for any candidate word, the selected candidate word is used as the new keyword.

[0193] For example, during browsing, a user might select a candidate word as a new keyword. This new keyword can then be aggregated with existing keywords, and new candidate words can be obtained based on the aggregation result. These new candidate words are then displayed in a second control. (Reference) Figure 4E , Figure 4E This is a schematic diagram of the search page provided in an embodiment of this application; compared to Figure 4C The original candidate word "what to eat" is selected, and the new candidate word corresponding to the new keyword 401E in the second control 402E.

[0194] In this embodiment, keywords can represent the user's search intent. Candidate words can be updated based on new keywords selected by the user, and more relevant information can be recommended to the user, making it easier for the user to flexibly select search terms to obtain search results that better meet their needs.

[0195] In step 605A, in response to a deletion operation for any keyword, the keyword is hidden.

[0196] For example, if a user is not satisfied with a keyword during browsing, they can delete it. The deleted keyword is hidden, and the corresponding candidate words are also hidden. The displayed candidate words are updated based on the deleted keyword.

[0197] In step 606A, candidate words related to the updated keywords are displayed based on the updated keywords.

[0198] Example, reference Figure 4F , Figure 4F This is a schematic diagram of a word cloud provided in an embodiment of this application; Figure 4FThe second control 402C includes three keywords, one of which is a new keyword. Based on these three keywords, the candidate words in the word cloud 401B are updated, resulting in multiple new candidate words 401F. (Continue to refer to...) Figure 4E , Figure 4E The candidate words in the second control 402E are compared to Figure 4C Unlike other keywords, candidate words are updated as the keywords are updated.

[0199] In some embodiments, during the execution of step 606A, if the user is not satisfied with the currently displayed keywords or candidate words, they can browse at any time to obtain other candidate words or keywords. The terminal device then executes steps 602A to 606A again to obtain words that better meet the user's needs.

[0200] In step 607A, in response to a trigger operation for any word, the search results are displayed.

[0201] Example, reference Figure 4A or Figure 4C , Figure 4E The search results are displayed in the third control, 403A. The search can be triggered and displayed only after the final search term is determined. If the user is satisfied with the current search term, they will begin browsing the results page; if not, they can reselect search terms, including deselecting some keywords, browsing, and adding new keywords. This process does not require any modification or deletion by the user in the input box. Alternatively, the keyword can be used as the default search term, and the search results corresponding to that keyword can be displayed before the search term is determined.

[0202] In some embodiments, if the user is not satisfied with the search results, they can browse the recommended keywords and candidate words in the second control again, select other search terms, and the terminal device will execute steps 602A to 607A again to obtain search results that better meet the user's needs.

[0203] The search processing method provided in this application eliminates the need for users to spend considerable time crafting descriptions of their search intent. By browsing the recommended word cloud, users can quickly discover a large number of relevant keywords and candidate words. These keywords can include associations in multiple directions related to the initial input and can be continuously reinforced or expanded upon in areas of user interest. Users can also quickly select search terms based on keywords and candidate words without needing to modify or delete them through the search box and keyboard, but rather through simple browsing and clicking. These innovative interactive processes significantly improve the efficiency of users expressing objective and accurate search terms.

[0204] This application's embodiments perform real-time understanding and recommendation based on user-input query information, providing a novel and convenient way for users to input their true search intent through interactive controls. This saves users significant time, effectively reducing the thinking and operational costs involved in finding their true search intent, greatly improving the efficiency of the process, and enhancing the accuracy of search results.

[0205] The following description continues to illustrate the exemplary structure of the search processing device 455 provided in the embodiments of this application as a software module. In some embodiments, such as Figure 2 As shown, the software modules stored in the search processing device 455 in the memory 450 may include: an input module 4551, for displaying query information entered through the information input operation in response to the information input operation; a display module 4552, for displaying multiple prompt words corresponding to the query information, wherein each prompt word is a separate option; and a search module 4553, for displaying search results corresponding to at least one prompt word selected by the selection operation in response to the selection operation.

[0206] In some embodiments, the display module 4552 is configured to display prompt words belonging to multiple types in multiple arrangement directions, wherein at least one type of prompt word is distributed in each arrangement direction.

[0207] In some embodiments, the multiple prompt words include: at least one keyword corresponding to the query information, and multiple candidate words related to the at least one keyword; the display module 4552 is used to display at least one keyword, wherein the at least one keyword is extracted from the query information or obtained by semantic understanding of the query information; in multiple arrangement directions starting from the at least one keyword, candidate words belonging to multiple types are displayed respectively, wherein at least one type of candidate words are distributed in each arrangement direction, and the different types of candidate words are all associated with at least one keyword.

[0208] In some embodiments, the following two parameters for each type of candidate word are positively correlated: the distance between the candidate word and at least one keyword, and the relevance of the candidate word to at least one keyword.

[0209] In some embodiments, in the display area formed around at least one keyword, the arrangement direction of each type satisfies one of the following conditions: the span of the angle interval corresponding to the arrangement direction of each type is the same, and the span is the ratio of the following two parameters: the span of the angle interval of the display area, and the number of candidate word types associated with at least one keyword.

[0210] In some embodiments, the display module 4552 is configured to display a delete button in each of the multiple arrangement directions when candidate words belonging to multiple types are displayed in multiple arrangement directions outward from at least one keyword; in response to a trigger operation of the first button in the first arrangement direction, for the candidate words of the first type corresponding to the first arrangement direction, perform one of the following processes: replace the candidate words of the first type with candidate words of the second type, wherein the second type is different from the multiple types; stop displaying the candidate words of the first type and update the angle range of other arrangement directions to cover the first arrangement direction, wherein the other arrangement directions are arrangement directions other than the first arrangement direction among the multiple arrangement directions.

[0211] In some embodiments, when candidate words belonging to multiple types are displayed in multiple arrangement directions outward from at least one keyword, the display module 4552 is configured to, in response to a browsing operation along the first arrangement direction, move the content distributed in the first arrangement direction along the first arrangement direction and display new candidate words of the first type in the free area formed by moving at least part of the content in the first arrangement direction, wherein the distance of movement is positively correlated with the magnitude of the browsing operation, and the first type is the type corresponding to the first arrangement direction.

[0212] In some embodiments, the display module 4552 is configured to determine new candidate words of the first type by: querying the word graph based on edge keywords and at least one keyword, searching for all nodes at the next level corresponding to the edge keywords and at least one keyword respectively in the word graph, wherein the edge keywords are candidate words located at the edge of the first arrangement direction before responding to the browsing operation; determining the co-occurrence count of all nodes at the next level of the edge keywords and at least one keyword, and sorting the nodes in descending order according to the co-occurrence count to obtain a descending order sorting result; deleting nodes with semantic opposites to the edge keywords and at least one keyword from the multiple nodes at the head of the descending order sorting result, and deleting the candidate words of the first type that have already been displayed; and taking the words represented by the remaining at least one node as new candidate words of the first type.

[0213] In some embodiments, the display module 4552 is configured to, if the browsing operation is not the first browsing operation received, and after displaying multiple prompts corresponding to the query information, in response to a browsing operation that crosses from a first arrangement direction to a second arrangement direction in the second control, display candidate words of a type corresponding to the second arrangement direction in the area between the current browsing position and at least one keyword.

[0214] In some embodiments, the display module 4552 is configured to distribute n candidate words of each type in n arrangement directions when the number n of multiple types is less than or equal to the type number threshold N; and to distribute N-1 candidate words of each type in the first N-1 arrangement directions and n-N+1 candidate words of each type in the Nth arrangement direction when the number n of multiple types is greater than the type number threshold N.

[0215] In some embodiments, the display module 4552 is used for at least one of the following display methods: applying different display parameters to candidate words distributed in different arrangement directions; displaying the boundary line between any two adjacent arrangement directions.

[0216] In some embodiments, the display module 4552 is used to display prompt words of multiple types in multiple queues, wherein the multiple queues are in the form of multiple rows or columns, and each queue contains prompt words of at least one type.

[0217] In some embodiments, the multiple prompt words include: at least one keyword corresponding to the query information, and multiple candidate words related to the at least one keyword; the display module 4552 is used to display at least one keyword, wherein the at least one keyword is extracted from the query information or obtained by semantic understanding of the query information; and candidate words belonging to multiple types are displayed in multiple queues, wherein at least one type of candidate words are distributed in each queue, and the different types of candidate words are all associated with at least one keyword.

[0218] In some embodiments, the display module 4552 is configured to, in response to a browsing operation along a target direction in a target queue, move the content in the target queue along the target direction and display a new candidate word of the target type in the free area formed by removing at least part of the content in the target queue when candidate words belonging to multiple types are displayed in multiple queues respectively, wherein the distance moved is positively correlated with the magnitude of the browsing operation, and the target type is the type corresponding to the target queue.

[0219] In some embodiments, when candidate words belonging to multiple types are displayed in multiple queues respectively, the display module 4552 is used to display a delete button on each queue; in response to a trigger operation on the first button on the first queue, for the candidate words of the first type corresponding to the first queue, one of the following processes is performed: replacing the candidate words of the first type with candidate words of the second type, wherein the second type is different from the multiple types; stopping the display of candidate words of the first type and increasing the size of other queues to cover the first queue, wherein the other queues are queues other than the first queue.

[0220] In some embodiments, the display module 4552 is configured to distribute n candidate words of each type in n queues when the number n of multiple types is less than or equal to the type number threshold N; and to distribute N-1 candidate words of each type in the first N-1 queues and n-N+1 candidate words of each type in the Nth queue when the number n of multiple types is greater than the type number threshold N.

[0221] In some embodiments, at least one keyword is selected by default, and multiple candidate words are unselected by default. Each selection operation is used to trigger the display of the corresponding search results. The display module 4552 is used to display that at least one first target word is selected in response to the selection operation for at least one first target word, and to display the search results corresponding to at least one keyword and at least one first target word, wherein at least one first target word is at least one candidate word selected by the selection operation among multiple candidate words.

[0222] In some embodiments, after displaying search results corresponding to at least one keyword and at least one first target word, the search module 4553 is configured to, in response to a cancellation operation for at least one second target word, display that at least one second target word is in an unselected state, wherein at least one second target word is at least one of at least one keyword and at least one first target word that has been cancelled by the cancellation operation; and display updated search results, wherein the updated search results correspond to at least one keyword and at least one first target word other than at least one second target word.

[0223] In some embodiments, multiple prompts are displayed in a second control, with at least one keyword located at the center of the second control.

[0224] In some embodiments, the search module 4553 is configured to display search results corresponding to at least one prompt word selected by the selection operation in response to a trigger operation on the search button in the first control.

[0225] In some embodiments, the display module 4552 is configured to display a search intent when multiple prompts corresponding to the query information are displayed, wherein the search intent represents the intention to search by inputting the query information.

[0226] In some embodiments, the display module 4552 is configured to determine multiple types of candidate words by: extracting at least one keyword from query information; when there are multiple keywords, searching for all nodes at the next level of each keyword in the word graph; determining the co-occurrence count of all nodes at the next level of multiple keywords, and sorting the nodes in descending order according to the co-occurrence count to obtain a descending order sorting result; deleting nodes with semantic opposites to any keyword from the multiple nodes at the head of the descending order sorting result, and using the words represented by the remaining multiple nodes as multiple types of candidate words.

[0227] In some embodiments, the display module 4552 is configured to determine multiple types of candidate words by: extracting at least one keyword from query information; when the number of at least one keyword is one, searching for all nodes at the next level of the keyword in the word graph; deleting nodes that are semantically opposite to any keyword from all nodes at the next level of the keyword, and using the words represented by the remaining multiple nodes as multiple types of candidate words.

[0228] This application provides a computer program product comprising a computer program or computer-executable instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer-executable instructions from the computer-readable storage medium and executes the computer-executable instructions, causing the computer device to perform the search processing method described above in this application.

[0229] This application provides a computer-readable storage medium storing computer-executable instructions. When these computer-executable instructions are executed by a processor, they cause the processor to execute the search processing method provided in this application. For example, ... Figure 3A The search processing method is shown.

[0230] In some embodiments, the computer-readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or it may be a variety of devices including one or any combination of the above-mentioned memories.

[0231] In some embodiments, computer-executable instructions may take the form of programs, software, software modules, scripts, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as stand-alone programs or as modules, components, subroutines, or other units suitable for use in a computing environment.

[0232] As an example, computer-executable instructions may, but do not necessarily, correspond to files in a file system. They may be stored as part of a file that holds other programs or data, for example, in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple co-located files (e.g., a file that stores one or more modules, subroutines, or code sections).

[0233] As an example, computer-executable instructions can be deployed to execute on a single electronic device, or on multiple electronic devices located at one location, or on multiple electronic devices distributed across multiple locations and interconnected via a communication network.

[0234] In summary, this application embodiment displays corresponding keywords and candidate words based on the query information input in the first control, and provides different keywords and candidate words as options based on the query information. This allows users to flexibly select one or more search terms according to their search intent, thereby ensuring that the search results obtained with the search terms match their search intent. This facilitates users in searching for the information they need, saves users' operation steps in the process of searching for information, and improves search efficiency and the accuracy of search results.

[0235] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, and improvements made within the spirit and scope of this application are included within the scope of protection of this application.

Claims

1. A search processing method, characterized in that, The method includes: In response to an information input operation, the query information entered through the information input operation is displayed; The system displays multiple suggested words corresponding to the query information, wherein each suggested word is treated as a separate option; the multiple suggested words include: at least one keyword corresponding to the query information, and multiple candidate words related to the at least one keyword; The at least one keyword is displayed, wherein the at least one keyword is extracted from the query information or obtained by semantic understanding of the query information; Candidate words belonging to multiple types are displayed in multiple queues, wherein the multiple queues are in the form of multiple rows or columns, and each queue contains at least one candidate word of the specified type, and candidate words of different types are all associated with the at least one keyword; In response to a browsing operation along a target direction in a target queue, the content in the target queue is moved along the target direction, and a new candidate word of the target type is displayed in the free area formed by removing at least part of the content in the target queue, wherein the distance of movement is positively correlated with the magnitude of the browsing operation, and the target type is the type corresponding to the target queue; In response to a selection operation, the search results corresponding to at least one of the suggested words selected by the selection operation are displayed.

2. The method according to claim 1, characterized in that, The multiple prompts corresponding to the query information are displayed, including: Multiple types of prompt words are displayed in multiple arrangement directions, wherein at least one prompt word of the type is distributed in each of the arrangement directions.

3. The method according to claim 2, characterized in that, The display of prompts belonging to multiple types in multiple arrangement directions includes: In multiple arrangement directions starting from the at least one keyword, candidate words belonging to multiple types are displayed respectively, wherein at least one candidate word of the type is distributed in each arrangement direction, and candidate words of different types are all associated with the at least one keyword.

4. The method according to claim 3, characterized in that, The following two parameters of each candidate word type are positively correlated: the distance between the candidate word and the at least one keyword, and the relevance of the candidate word to the at least one keyword.

5. The method according to claim 3, characterized in that, In the display area formed around the at least one keyword, the arrangement direction of each type satisfies one of the following conditions: The span of the angle interval corresponding to each of the types of arrangement is the same, and the span is the ratio of the following two parameters: the span of the angle interval of the display area, and the number of candidate word types related to the at least one keyword.

6. The method according to claim 3, characterized in that, When candidate words belonging to multiple types are displayed in multiple arrangement directions emanating from the at least one keyword, the method further includes: A delete button is displayed for each of the aforementioned arrangement directions; In response to a trigger operation of a first button in a first arrangement direction, for the candidate word of the first type corresponding to the first arrangement direction, one of the following processes is performed: Replace the candidate words of the first type with the candidate words of the second type, wherein the second type is distinct from the plurality of types; Stop displaying the candidate words of the first type and update the angle range of other arrangement directions to cover the first arrangement direction, wherein the other arrangement directions are arrangement directions other than the first arrangement direction among the plurality of arrangement directions.

7. The method according to claim 3, characterized in that, When candidate words belonging to multiple types are displayed in multiple arrangement directions emanating from the at least one keyword, the method further includes: In response to a browsing operation along a first arrangement direction, the content distributed along the first arrangement direction is moved along the first arrangement direction, and In the free area formed by moving at least part of the content out in the first arrangement direction, new candidate words of the first type are displayed, wherein the distance of the movement is positively correlated with the magnitude of the browsing operation, and the first type is the type corresponding to the first arrangement direction.

8. The method according to claim 7, characterized in that, The method further includes: The first type of new candidate words are determined in the following way: The word graph is queried based on the edge keywords and the at least one keyword. All nodes of the next level corresponding to the edge keywords and the at least one keyword are found in the word graph. The edge keywords are the candidate words located at the edge of the first arrangement direction before responding to the browsing operation. Determine the co-occurrence count of the edge keyword and all nodes at the next level of the at least one keyword, and sort the nodes in descending order according to the co-occurrence count to obtain the descending order sorting result; In the descending sort result, among the multiple nodes at the head, delete the nodes that are semantically opposite to the edge keyword and the at least one keyword, and delete the candidate words of the first type that have already been displayed; The word represented by at least one of the remaining nodes is used as a new candidate word of the first type.

9. The method according to claim 3, characterized in that, If the browsing operation is not the first browsing operation received, and after displaying multiple prompts corresponding to the query information, the method further includes: In response to a browsing operation that crosses from a first arrangement direction to a second arrangement direction in a second control, the candidate words of the type corresponding to the second arrangement direction are displayed in the area between the current browsing position and the at least one keyword.

10. The method according to any one of claims 3 to 9, characterized in that, The distribution of at least one candidate word of the aforementioned type in each of the aforementioned arrangement directions includes: When the number n of the multiple types is less than or equal to the type number threshold N, there are n candidate words of the type distributed one-to-one in the n arrangement directions; When the number n of the multiple types is greater than the type number threshold N, there are N-1 candidate words of the type distributed one-to-one in the first N-1 arrangement directions, and there are n-N+1 candidate words of the type distributed in the Nth arrangement direction.

11. The method according to claim 2, characterized in that, The method further includes: At least one of the following display methods: Different display parameters are applied to candidate words distributed in different arrangement directions; Display the dividing line between any two adjacent arrangement directions.

12. The method according to claim 1, characterized in that, When displaying candidate words belonging to multiple types in multiple queues, the method further includes: A delete button is displayed on each of the queues; In response to a trigger operation on a first button in the first queue, for the candidate word of the first type corresponding to the first queue, one of the following processes is performed: Replace the candidate words of the first type with the candidate words of the second type, wherein the second type is distinct from the plurality of types; Stop displaying the candidate words of the first type and increase the size of other queues to cover the first queue, wherein the other queues are the queues other than the first queue.

13. The method according to any one of claims 1 and 12, characterized in that, The distribution of at least one candidate word of the aforementioned type in each queue includes: When the number n of the multiple types is less than or equal to the type number threshold N, there are n candidate words of the type distributed one-to-one in the n queues; When the number n of the multiple types is greater than the type number threshold N, there are N-1 candidate words of the type distributed one-to-one in the first N-1 queues, and there are n-N+1 candidate words of the type distributed in the Nth queue.

14. The method according to any one of claims 1 to 9, 11 to 12, characterized in that, The at least one keyword is selected by default, and the multiple candidate words are unselected by default. Each selection operation is used to trigger the display of the corresponding search results. The step of responding to a selection operation by displaying search results corresponding to at least one of the prompt words selected by the selection operation includes: In response to a selection operation for at least one first target word, the at least one first target word is displayed as selected, and the search results corresponding to the at least one keyword and the at least one first target word are displayed, wherein the at least one first target word is at least one of the candidate words selected by the selection operation from the plurality of candidate words.

15. The method according to claim 14, characterized in that, After displaying the search results corresponding to the at least one keyword and the at least one first target term, the method further includes: In response to a cancellation operation for at least one second target word, the at least one second target word is displayed as unselected, wherein the at least one second target word is at least one of the at least one keyword and the at least one first target word that was cancelled by the cancellation operation; Display updated search results, wherein the updated search results correspond to at least one keyword and at least one first target word, in addition to the at least one second target word.

16. The method according to any one of claims 1, 3 to 9, 11 to 12, characterized in that, The first control for responding to the information input operation includes a search box and a search button, wherein the search box is used to input the query information; Displaying the search results corresponding to at least one of the prompt words selected by the selection operation includes: In response to a trigger operation on the search button in the first control, the search results corresponding to at least one of the prompt words selected by the selection operation are displayed.

17. The method according to any one of claims 1 to 9, 11 to 12, characterized in that, The method further includes: The candidate words of the multiple types are determined in the following manner: Extract the at least one keyword from the query information; When there are multiple keywords, search all nodes at the next level of each keyword in the word graph; Determine the co-occurrence count of all nodes at the next level of the multiple keywords, and sort the nodes in descending order according to the co-occurrence count to obtain the descending order sorting result; In the descending sort result, among the multiple nodes at the head, delete the node whose semantics are opposite to any of the keywords, and use the words represented by the remaining multiple nodes as candidate words of multiple types.

18. The method according to any one of claims 1 to 9, 11 to 12, characterized in that, The method further includes: The candidate words of the multiple types are determined in the following manner: Extract the at least one keyword from the query information; When the number of the at least one keyword is one, search for all nodes at the next level of the keyword in the word graph; From all nodes at the next level of the keyword, delete nodes whose semantics are opposite to any of the keywords, and use the words represented by the remaining nodes as candidate words of multiple types.

19. A search processing apparatus, characterized in that, The device includes: An input module is used to respond to an information input operation and display the query information entered through the information input operation; The display module is used to display multiple prompt words corresponding to the query information, wherein each prompt word is a separate option; the multiple prompt words include: at least one keyword corresponding to the query information, and multiple candidate words related to the at least one keyword; displaying the at least one keyword, wherein the at least one keyword is extracted from the query information or obtained by semantic understanding of the query information; displaying candidate words belonging to multiple types in multiple queues, wherein the multiple queues are in the form of multiple rows or columns, and each queue contains at least one candidate word of the specified type, and candidate words of different types are all associated with the at least one keyword; in response to a browsing operation along a target direction in the target queue, moving the content in the target queue along the target direction, and removing at least a portion of the content from the target queue to form a free area, displaying new candidate words of the target type, wherein the distance of movement is positively correlated with the magnitude of the browsing operation, and the target type is the type corresponding to the target queue; The search module is configured to, in response to a selection operation, display search results corresponding to at least one of the prompt words selected by the selection operation.

20. The apparatus according to claim 19, characterized in that, The display module is also used to display prompt words of multiple types in multiple arrangement directions, wherein at least one prompt word of the type is distributed in each of the arrangement directions.

21. The apparatus according to claim 20, characterized in that, The display module is further configured to display candidate words of multiple types in multiple arrangement directions outward from the at least one keyword, wherein at least one candidate word of the type is distributed in each arrangement direction, and candidate words of different types are all associated with the at least one keyword.

22. The apparatus according to claim 21, characterized in that, The following two parameters of each candidate word type are positively correlated: the distance between the candidate word and the at least one keyword, and the relevance of the candidate word to the at least one keyword.

23. An electronic device, characterized in that, The electronic device includes: Memory is used to store executable instructions for a computer; A processor, when executing computer-executable instructions or computer programs stored in the memory, implements the search processing method according to any one of claims 1 to 18.

24. A computer-readable storage medium storing computer-executable instructions or a computer program, characterized in that, When the computer-executable instructions are executed by a processor, they implement the search processing method according to any one of claims 1 to 18.

25. A computer program product comprising a computer program or computer-executable instructions, characterized in that, When the computer program or computer-executable instructions are executed by a processor, the search processing method according to any one of claims 1 to 18 is implemented.

Images