Information processing method, information processing device, computer device, and computer program

Abstract

An embodiment of the present application provides an information processing method and related devices, the method including: displaying an information search interface, the information search interface being used to perform information searches based on search fields; displaying subscription fields, the subscription fields referring to configured search fields; and outputting attention information related to the subscription fields. The embodiment of the present application can enrich the information search method, improve the flexibility of information search, and better adapt to situations in which a subject continuously pays attention to information.

Description

[Technical Field]

[0001] This application claims priority to a Chinese patent application filed with the China Patent Office on August 23, 2023, bearing application number 202311070020.1 and entitled "Information Processing Method and Related Apparatus," the entire contents of which are incorporated herein by reference.

[0002] The present application relates to the technical field of the Internet, and in particular to information processing methods and related devices, and more particularly to information processing methods, information processing devices, computer devices, computer-readable storage media, and computer program products. [Background technology]

[0003] Currently, a subject (e.g., a user) can input a search field (query) through an information search interface provided by a client terminal, thereby searching for information related to the search field. The information search interface can also record the subject's search history, and when the subject re-enters the information search interface, a recent history search field (e.g., the last few days) can be displayed. The subject can tap the history search field to start a re-search in the history search field to obtain task recall results. Practice has shown that existing information searches are too monotonous and not flexible enough to adapt to situations where the subject continuously focuses on information. Summary of the Invention [Means for solving the problem]

[0004] The embodiments of the present application provide an information processing method and related devices, which can enrich information retrieval methods, improve the flexibility of information retrieval, and better adapt to scenes where subjects continuously pay attention to information.

[0005] In one aspect, embodiments of the present application provide a method of processing information, the method comprising: displaying an information search interface, the information search interface being used to conduct an information search based on a search field; displaying a subscription field, the subscription field referring to a configured search field; and outputting attention information related to the subscription field.

[0006] In one aspect, an embodiment of the present application provides an information processing device, the device including a display unit and a processing unit; The display unit is used to display an information search interface, and the information search interface is used to perform information search according to a search field; The display unit is further used to display a subscription field, and the subscription field refers to a configured search field; The processing unit is used for outputting attention information related to a subscription field.

[0007] In one aspect, embodiments of the present application provide a computer device, the computer device comprising: a processor adapted to execute computer programs; and a computer-readable storage medium having a computer program stored thereon, the computer program implementing the information processing method when executed by a processor.

[0008] In one aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored therein, the computer program being loaded into a processor and executing the information processing method.

[0009] In one aspect, an embodiment of the present application provides a computer program product, the computer program product including a computer program or computer instructions that, when executed by a processor, implements the information processing method.

[0010] In an embodiment of the present application, an information search interface is displayed, and the information search interface is used to perform information search based on search fields. From this, it can be seen that by displaying an information search interface, information search based on search fields can be realized. A subscription field is displayed, and the subscription field refers to a configured search field. From this, it can be seen that a subject can configure a search field as a subscription field according to their continuous attention needs. By displaying the subscription field, the configured search fields that need continuous attention can be presented, and the subject can know the search fields that they are continuously interested in. Attention information related to the subscription field is output. In this way, information search can be continuously performed according to the subscription field to obtain and output information that the subject is continuously interested in, which enriches the information search method and at the same time improves the flexibility of information search, thereby better adapting to the scenarios in which the subject continuously pays attention to information. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is an architecture diagram of an information processing system provided by one exemplary embodiment of the present application. [Figure 2a] 1 is a schematic diagram of an information processing process provided by one exemplary embodiment of the present application. [Figure 2b] FIG. 2 is a flow diagram of a subscription renewal notification provided by one exemplary embodiment of the present application. [Figure 3a]1 is a flow diagram of an information processing method provided by one exemplary embodiment of the present application. [Figure 3b] FIG. 1 is a schematic diagram of outputting attention information related to a subscription field provided by one exemplary embodiment of the present application; [Figure 4a] FIG. 2 is a schematic diagram showing N subscription fields provided by one exemplary embodiment of the present application; [Figure 4b] FIG. 10 is a schematic diagram showing N subscription fields provided by another exemplary embodiment of the present application; [Figure 4c] FIG. 10 is a schematic diagram showing N subscription fields provided by another exemplary embodiment of the present application; [Figure 4d] FIG. 1 is a schematic diagram of a collapsed display of N subscription fields provided by one exemplary embodiment of the present application; [Figure 5a] FIG. 1 is a schematic diagram of outputting multiple documents that match subscription fields in a fixed area provided by one exemplary embodiment of the present application; [Figure 5b] FIG. 1 is a schematic diagram of outputting a content update notification in a floating area provided by an exemplary embodiment of the present application; [Figure 6a] FIG. 1 is a schematic diagram of setting subscription fields provided by one exemplary embodiment of the present application; [Figure 6b] FIG. 10 is a schematic diagram of setting a subscription field provided by another exemplary embodiment of the present application; [Figure 6c] FIG. 10 is a schematic diagram of setting subscription fields provided by yet another exemplary embodiment of the present application; [Figure 6d] FIG. 10 is a schematic diagram of setting subscription fields provided by another exemplary embodiment of the present application; [Figure 6e] FIG. 1 is a schematic diagram of setting subscription fields provided by one exemplary embodiment of the present application; [Figure 7a] FIG. 1 is a schematic diagram of deleting words in a subscription field provided by one exemplary embodiment of the present application; [Figure 7b] FIG. 2 is a schematic diagram showing multiple documents matching a subscription provided by one exemplary embodiment of the present application. [Figure 7c] FIG. 1 is a schematic diagram of a side-by-side display of multiple documents provided by one exemplary embodiment of the present application. [Figure 7d] 1 is a schematic diagram of clustering and displaying multiple documents provided by one exemplary embodiment of the present application; [Figure 7e] FIG. 2 is a schematic diagram illustrating the display of viewed and unviewed documents in different areas of an information retrieval interface provided by one exemplary embodiment of the present application. [Figure 7f] FIG. 2 is a schematic diagram of a display of viewed and unviewed documents provided by one exemplary embodiment of the present application; [Figure 8a] FIG. 2 is a schematic diagram of a content update notification provided by one exemplary embodiment of the present application. [Figure 8b] FIG. 10 is a schematic diagram of a content update notification provided by another exemplary embodiment of the present application; [Figure 9] 1 is a structural schematic diagram of an information processing device provided by an exemplary embodiment of the present application; [Figure 10] FIG. 1 is a structural schematic diagram of a computer device provided by one exemplary embodiment of the present application. DETAILED DESCRIPTION OF THE INVENTION

[0012] First, technical terms related to the embodiments of the present application will be introduced.

[0013] 1. Search field and subscription field

[0014] A search field (query) refers to a string of characters used to search for information, and the search field may include keywords, for example, a person's name, a dish, a game name, etc., or the search field may include a sentence, for example, a search field may include "What dishes are available in XX city?", and the embodiments of the present application are not limited thereto.

[0015] A subscription field refers to a configured search field that requires continuous attention (i.e., that requires attention within a certain period of time). Configuring a subscription field can be understood as performing attention processing on a search field that requires continuous attention. That is, configuring a search field as a subscription field can trigger continuous attention to attention information related to the subscription field, thereby providing a kind of information search service that can be continuously or periodically updated for the target.

[0016] 2. Important Information

[0017] The term "attention information" refers to information related to a subscription field. The attention information can be understood as information of continuous interest to a subject, obtained by constantly (real-time or periodically) searching for information using the subscription field. The attention information includes, but is not limited to, business recall results obtained by searching for information according to the subscription field, content update notifications related to the subscription field, etc. Here, the business recall results refer to the results of recalling or extracting some document by taking action (e.g., performing an information search) for a specific business need (e.g., for the information search need of the subscription field). The business recall results may include multiple documents that match the subscription field. Matching here refers to a relatively high degree of match between the document content and the subscription field (e.g., higher than a preset threshold). The multiple documents may include search documents and / or recommended documents. The so-called search documents refer to documents that match the subscription field searched by the search recall link, and the so-called recommended documents refer to documents associated with a topic or tag mapped to the subscription field searched by the recommendation recall link. In the present embodiment, the function of the search recall link is mainly to directly search for information according to the subscription field. The function of the recommendation recall link is mainly to first perform a mapping match on the subscription field to obtain the topic or tag mapped to the subscription field, and then perform information search based on the mapped topic or tag. The content update notification related to the subscription field is used to prompt that there is an update to the information related to the subscription field (i.e., the featured information). The content update notification may include at least one of the number of updated documents and the recommendation reason. Here, the number of updated documents refers to the number of documents that have been updated. For example, the number of updated documents is 3, which means that there are three updated documents.The recommendation reason may include the document content type of the updated document. For example, if the subscription field includes "model," the document content type of the updated document may be the technical principle of the model, the model introduction, etc., and the recommendation reason may be "a document update exists to describe the technical principle of the model," or "a document update exists to describe a new model and related introduction."

[0018] 3. Document Topics and Tags

[0019] A document may include text, video, audio, images, web pages, etc., and the present application does not impose any limitations thereon. A document topic is used to describe the main content of the entire document. For example, the document topic may be the title of the text and the name of the video. A document tag is used to describe key information of the document. The key information may include type, person, time, place, etc. Here, the type may be, for example, game, food, model, etc. For example, if the document is a video, for a video titled "Have you ever eaten this type of XX food at XX?", the topic of the video may be "Have you ever eaten this type of XX food at XX?", and the tags of the video may be, for example, XX place and XX type food. For example, for a sentence titled "Xiaoming is good at playing XX games," the topic of the sentence may be "Xiaoming is good at playing XX type games," and the tags of the sentence may include Xiaoming and XX type games.

[0020] It is particularly important to note that in this application, relevant data in the relevant information processing process may include, for example, a subject's behavior log (e.g., history search fields, document browsing history, and accounts the subject is interested in, etc.), subject attribute information (e.g., nickname and region), etc. When embodiments of this application are implemented in a specific product or technology, subject permission or consent must be obtained, and the collection, use, and processing of relevant data must comply with relevant laws, regulations, and standards, meet the principles of lawfulness, legitimacy, and necessity, and not involve the acquisition of data types prohibited or restricted by laws or regulations. In some optional embodiments, relevant data related to embodiments of this application may be acquired after the subject's independent consent is obtained, and the use of the relevant data may be clearly indicated to the subject upon obtaining the subject's independent consent.

[0021] An embodiment of the present application provides an information processing solution, the general principle of which is as follows: an information search interface is displayed, the information search interface is used to perform information search based on search fields, and subscription fields are displayed, where the subscription fields refer to configured search fields, and attention information related to the subscription fields is output. In this way, not only can information search be performed based on the search fields in the information search interface, but also the subject can configure the search fields as subscription fields according to their continuous attention needs, and by displaying the subscription fields, the configured search fields that need to be continuously paid attention can be presented, and the subject can know the search fields that are continuously interested. Furthermore, by continuously performing information search according to the subscription fields, the information that the subject is continuously interested in can be obtained and output, which enriches the information search method and at the same time improves the flexibility of information search and can better adapt to the scenarios in which the subject continuously pays attention to information.

[0022] Next, a related discussion will be given for the information processing system provided by the embodiment of the present application.

[0023] Referring to FIG. 1, an architecture diagram of an information processing system provided by an exemplary embodiment of the present application is shown. The information processing system includes terminal 101, terminal 102, ..., and more terminals, and the present application does not limit the number of terminals. The information processing system further includes server 103, and naturally, the number of servers may also be multiple, and the present application still does not limit this. Here, a terminal can be understood as a terminal used by a target to view a document. Here, any terminal in the information processing system and server 103 can be directly or indirectly connected via wired or wireless communication, and information can be exchanged between any two terminals via server 103. The following description will be given taking one of terminals 101 as an example.

[0024] The terminal 101 may provide an information search interface, allowing any subject using the terminal 101 to enter a search field in the information search interface to perform an information search, and output information matching the search field through the information search interface. The matching information may be a search result obtained by performing an information search using the search field. In addition, in an embodiment of the present application, the subject may further configure a subscription field in the information search interface, where the subscription field refers to the configured search field that requires continuous attention. In this case, the terminal 101 may output attention information related to the subscription field. The terminal 101 may be a smartphone, a tablet computer, a laptop computer, a desktop computer, a smart vehicle, a smart wearable device, etc., and the embodiment of the present application is not limited thereto.

[0025] The server 103 is used to provide information processing services for the terminal 101. The information processing services here may include, but are not limited to, services such as performing information search based on search fields, providing subscription field customization services for targets, performing information search based on subscription fields, etc. The server 103 may be an independent physical server, a server cluster or a distributed system consisting of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), big data, and artificial intelligence platforms.

[0026] Referring to Figure 2a, this is a schematic diagram of an information processing process provided by an embodiment of the present application. The information processing process can be executed by the server 103, and includes: 1) offline behavior log and document content understanding. This part includes candidate search field recommendation calculation, target information calculation, and document content understanding. 2) subscription field recommendation and customization. 3) subscription field search (i.e., information search based on subscription field). 4) subscription search result feedback. 5) content update notification. The above steps will be expanded and described below.

[0027] 1) Offline behavior log and document content understanding

[0028] 1.1 Recommendation calculation for candidate search fields

[0029] (1) Collecting a target's behavior log for each time period in an offline state. Here, the time period can be flexibly set, for example, in days, such as 1 day, 2 days, etc. As another example, the time period can be set in hours, such as 1 hour, 2 hours, etc. As another example, the time period can be set in weeks, such as 1 week, 2 weeks, etc. The offline state refers to a state in which the device used by the target is disconnected from the Internet or network. The target's behavior log includes, but is not limited to, information search history (e.g., the target searches for information by entering a history search field within a history time period), video playback history (e.g., the history of videos played by the target), followed accounts (e.g., accounts that post a certain article or video that the target is following), likes (e.g., the target likes a certain article, a certain video, a certain video account, etc.), and sharing / forwarding by the target (e.g., the target shares and forwards a certain article and a certain video). After acquiring the target behavior log, a content tag clustering technique can be used to calculate candidate search fields for the behavior log, thereby obtaining one or more candidate search fields. Content tag clustering is a technique commonly used in the field of information retrieval, used to cluster documents with similar topics or content together, thereby enabling the content of a large number of documents to be better organized and displayed. In one implementation, the processing logic for using the content tag clustering technique includes: performing a word segmentation process on the content in the target behavior log (e.g., a history search field and a title of a history viewed document, etc.) to obtain one or more segmented words; vectorizing the one or more segmented words to obtain vectors for each segmented word; and performing a clustering calculation on the vectors of each segmented word to obtain tags corresponding to each segmented word, and setting the tags corresponding to each segmented word as candidate search fields.

[0030] (2) After obtaining the candidate search fields, the candidate search fields can be bound to the objects and stored in a database. Binding a candidate search field to an object refers to associating the candidate search field with the object identifier of the object (e.g., storing the candidate search field in a database). For example, the database may include a key-value database (abbreviated as KV database). In this case, the object identifier of the object can be stored as a key and the candidate search field as a value in the database. The object identifier here can be a nickname, an account, etc.

[0031] It should be understood that in the embodiments of the present application, the behavior log of each subject is collected, and the candidate search fields corresponding to each subject are determined according to the above method, and each subject can be bound to the corresponding candidate search field. In this case, when a subject configures a subscription field, the server can directly recommend the corresponding candidate search field for the subject, so that the subject can select it, thus not only improving the configuration efficiency of the subscription field, but also providing individual subscription services for the search field for each subject.

[0032] 1.2 Target Information Calculation

[0033] (1) As described in 1.1, content tag clustering technology can be used to perform content tag clustering processing on collected target behavior logs to obtain one or more tags, and in this case, the one or more tags can be considered target information. The so-called target information refers to information used to describe the image and characteristics of a target, which is a typical target character that is hypothesized (abstracted) from the target's actual behavior. It can be used to describe common characteristics of a type of target, which may include, for example, basic attributes (e.g., region), interests, hobbies, etc. For example, if the time period is set on a daily basis, i.e., if target behavior logs are collected on a daily basis (e.g., one day), content tag clustering technology can be used to perform content tag clustering processing on the collected target behavior logs to obtain one or more tags, and the one or more tags can be determined as target information for the target on that daily basis.

[0034] (2) Collect target behavior logs over multiple time periods (e.g., multiple days), and generate target information for multiple time periods according to (1). For example, if the time period is one day, content tag clustering processing can be performed on target behavior logs collected over multiple days (e.g., 100 days) to obtain target information for the multiple days (e.g., 100 days). The target information over multiple time periods and the target's attribute information are combined to obtain long-term target information for the target. So-called long-term target information refers to target information that does not change within a certain period of time (e.g., one or two years). The target's attribute information may include, but is not limited to, gender, region, nickname, etc.

[0035] (3) Obtaining target information and behavior logs of a target time period, and performing content tag clustering on the target information and behavior logs of the target time period to obtain the real-time interest needs of the target. The real-time interest needs are used to indicate the target's attention direction within the target time period, and the target time period may be the last three days, the last three hours, etc. In the present embodiment, without any limitation, "recent" refers to the system time when the target information and behavior logs of the target time period are obtained, i.e., the last three days refers to the three days prior to the system time, and the last three hours refers to the three hours prior to the system time. For example, obtaining target information and behavior logs of a target within the last three days, and the behavior logs may record, for example, what entertainment news the target has viewed within the last three days, and performing content tag clustering on the target information and behavior logs of the target within the last three days to obtain the target's real-time interest needs. The real-time interest needs are used to indicate that the target has recently paid relatively more attention to entertainment information.

[0036] It should be understood that both long-term target information and real-time interest needs can be applied in the process of information retrieval based on subscription fields, and serve as the basis for screening and sorting the retrieved information (e.g., retrieved documents).

[0037] 1.3. Document understanding

[0038] This part includes establishing correspondence between viewed documents and objects (i.e., generating a historical viewing record of the object), and building a search document library and a recommendation recall document library.

[0039] 1.3.1) Generate a historical browsing record of the subject.

[0040] A target's behavior log may include a document browsing history, which refers to a record of documents previously viewed by the target. For example, the document browsing history records when and what documents the target previously viewed, and these previously viewed documents may be referred to as previously viewed documents. That is, the document browsing history includes previously viewed documents. In an embodiment of the present application, a fingerprint calculation is performed on the previously viewed documents to obtain fingerprint features corresponding to the previously viewed documents. Fingerprint calculation is a technology used to determine the uniqueness of document content. By performing fingerprint calculation on a document, a unique identifier (fingerprint feature) can be generated to represent the document content. That is, the fingerprint feature of a document can be used to uniquely identify a document, and the fingerprint feature of a previously viewed document can be used to find the corresponding previously viewed document. In one implementation, the process of performing fingerprint calculation on the viewed document and obtaining fingerprint features corresponding to the viewed document may include: extracting a plurality of content features from the viewed document (here, the content features may include, but are not limited to, keywords in the document, sentences in the document, etc.); employing a hash algorithm to perform hash calculation on the plurality of content features to obtain hash values ​​corresponding to each content feature; and performing a merge process on the hash values ​​corresponding to each content feature to obtain fingerprint features of the viewed document. It is worth noting that the following related fingerprint calculations can all refer to the fingerprint calculation process of the viewed document.

[0041] After obtaining the fingerprint features corresponding to the viewed documents, one implementation can be to use the viewed documents together with the corresponding fingerprint features as a historical browsing record of the object, and to build a viewed document filter based on the historical browsing record. For example, the viewed document filter can be a Bloom filter that uses the length of the viewed document list as its size. Another implementation can be to write the viewed documents together with the corresponding fingerprint features as a historical browsing record of the object in a database. For example, if the database is a KV database, the object identifier of the object can be written as a key, and the viewed document and its corresponding fingerprint features can be written as a value in the database. Two examples are described in detail below.

[0042] Example 1: Assume that the viewed documents include viewed sentences, and the viewed document filters include viewed sentence filters. The process of generating a historical browsing record of an object for the viewed sentences may include: performing a fingerprint calculation on the body of the viewed sentences, and constructing a viewed sentence filter (e.g., the viewed sentence filter may be a Bloom filter using the length of the viewed sentence list as its size) based on the fingerprint features corresponding to the viewed sentences. Alternatively, assume that the database includes a KV database of viewed sentences, and the process of generating a historical browsing record of an object for the viewed sentences may include constructing a KV database of the viewed sentences using the object identifier of the object as a key and the fingerprint features of the viewed sentence as a value.

[0043] Example 2: Assume that the viewed document includes a viewed video, and the viewed document filter includes a viewed video filter. The process of generating a historical viewing record of the object for the viewed video can include: extracting frames from the viewed video at a preset rate (the preset rate can be set according to needs, for example, extracting one frame every 2 seconds); performing fingerprint calculation on the extracted video frames to obtain fingerprint features corresponding to the video frames; and constructing a viewed video filter (exemplary, the viewed video filter is a Bloom filter whose size is the length of the viewed video list) based on the fingerprint features corresponding to the video frames. Alternatively, assume that the database includes a KV database of viewed videos, and the process of generating a historical viewing record of the object for the viewed video can include constructing a KV database of viewed videos using the object identifier of the object as a key and the fingerprint features of the viewed video as a value.

[0044] It should be understood that the generated historical browsing record of the subject can be applied in the process of performing information search based on the subscription field, and can serve as the basis for filtering documents that match the subscription field, thereby preventing the subject from receiving documents that have already been viewed, avoiding redundant recommendations, and reducing the pressure on information exchange between the server and the terminal.

[0045] 1.3.2) Building a searchable document library.

[0046] In one implementation, global documents in the search document library can be directly retrieved, where global documents refer to all documents in the search document library. A fingerprint calculation is performed on each retrieved document to obtain the fingerprint features of each document. Each retrieved document is then selected as a candidate search document, and the fingerprint features of each document are written to the search document library as the document features of the corresponding candidate search document. Here, the search document library refers to a database that provides data support for information retrieval. That is, in this embodiment, information retrieval is performed in the search document library. The search document library can contain various types of documents, including, but not limited to, pre-defined documents, received documents, documents uploaded by the subject, or documents retrieved from the Internet, and the search document library supports continuous updates.

[0047] In another implementation, the search document library can be divided into a global search document library (gob) and a local search document library according to the document publication time, i.e., the search document library includes a global search document library (gob) and a local search document library. The document publication time refers to the time when a document is officially posted or published. The global search document library includes the global documents in the search document library and the fingerprint features of each document. In this case, the construction method of the global search document library includes: obtaining the global documents in the search document library, performing fingerprint calculation on each obtained document to obtain the fingerprint features of each document, and taking each obtained document as a candidate search document, and using the fingerprint features of each document as the document features of the corresponding candidate search document; and constructing a global search document library based on the candidate search documents and the document features of the candidate search documents.

[0048] The local search document library is used to store local documents, which are a relative concept of global documents, i.e., local documents refer to a portion of global documents. In this embodiment, local documents refer to documents whose document publication times are within a second preset time period in the search document library, and the second preset time period may be a time period corresponding to a week, hour, or day. For example, local documents may refer to documents posted within a certain week or a certain two weeks. The method for constructing the local search document library includes: obtaining documents whose document publication times are within the second preset time period, performing fingerprint calculation on each document within the second preset time period to obtain fingerprint features of each document within the second preset time period, and treating each document within the second preset time period as local search documents, and using the fingerprint features of each document within the second preset time period as document features of the corresponding local search documents; and constructing a local search document library based on the local search documents and the document features of the local search documents.

[0049] Here, the local search document library includes one or more of a new search document library (FOB) and a latest search document library (latest ob). In an embodiment of the present application, the various local search document libraries (i.e., the new search document library (FOB) and the latest search document library (latest ob)) can be constructed according to their respective time periods. For example, the new search document library can be constructed according to a weekly time period. For example, the new search document library can include documents posted within the last week or documents posted within the last two weeks. The latest search document library can be constructed according to a daily time period or an hourly time period. For example, the latest search document library can include documents posted within the last day or documents posted within the last hour. By constructing the various obs according to their respective time periods, search documents that match subscription fields can be easily obtained from the corresponding obs based on different needs, thereby improving information search efficiency. When the new search document library (fob) is constructed using a weekly time period (e.g., one week), it can acquire documents posted within the last week, perform fingerprint calculations on the acquired documents, and write the calculated fingerprint features as document features of the acquired documents and the corresponding document features into the new search document library (fob).When the latest search document library (latest ob) is constructed using a daily time period (e.g., one day), it can acquire documents posted within the last day, perform fingerprint calculations on the acquired documents, and write the calculated fingerprint features as document features of the acquired documents and the corresponding document features into the latest search document library (latest ob).

[0050] 1.3.3) Building a recommendation recall document library: Obtain documents from a certain period of time (e.g., the past month) to obtain P candidate recommendation documents, where P is a positive integer. Perform offline scoring on the P candidate recommendation documents according to recommendation influence factors, obtain a score for each of the P candidate recommendation documents, and sort the P candidate recommendation documents according to the highest score to obtain P sorted candidate recommendation documents. Obtain the top Y candidate recommendation documents from the sorted P candidate recommendation documents, and build a recommendation recall document library based on the top Y candidate recommendation documents. Here, the recommendation recall document refers to a database that provides data support for a target to make information recommendations. That is, in this embodiment, information retrieval and recommendation are performed in the recommendation recall document library based on subscription fields. The recommendation recall document library can contain a variety of relatively excellent documents, including, but not limited to, pre-defined documents, received documents uploaded by the target, and documents obtained from the Internet. The recommendation recall document library supports continuous updates. A so-called good document generally refers to a document that is of high quality (the quality of the text is higher than a pre-set quality threshold), complete, clear, accurate, and reliable (i.e., the content source is reliable, has clear evidence or data support, and can be relied upon).

[0051] The recommendation influence factors may include at least one of factors such as document originality, document activity, the number of likes of the document, the number of document shares, the number of document forwardings, the quality of the document body, and the time of document publication, etc. The recommendation influence factors can make the recommended recall documents include some relatively excellent documents.

[0052] (1) Document originality: refers to an original document that is independently created and does not refer to other documents. In an embodiment of the present application, offline scoring of P candidate recommendation documents according to recommendation influence factors and obtaining a score corresponding to each of the P candidate recommendation documents may include scoring the document originality of the input P candidate recommendation documents using an originality assessment model, obtaining a document originality score for each of the P candidate recommendation documents, and using the document originality score of each candidate recommendation document as the score of each candidate recommendation document. An originality assessment model refers to a model or system used to evaluate the originality of a work or document. The originality assessment model typically uses natural language processing technology and machine learning algorithms to determine the degree of document originality by comparing the similarity of the document with existing works and any possible plagiarism. For example, the originality assessment model may include, but is not limited to, a model based on TF-DF (word frequency and inverse document frequency), an N-gram (word fragment) model, a word embedding model, and a convolutional neural network model, and the present application is not limited thereto. (2) Document activity: This may refer to the frequency with which a document is viewed. The more frequently a candidate recommended document is viewed, the higher the activity of the candidate recommended document, and the higher the activity of the candidate recommended document, the higher the score of the candidate recommended document. (3) Number of document likes: This refers to the number of times a document has been liked. Liking is a type of social interaction behavior and refers to a way of expressing agreement, support, or preference for the content of a document. The higher the number of likes a candidate recommended document has, the higher the score of the candidate recommended document. (4) Number of document shares: This refers to the number of times a document has been shared. Sharing refers to the process of disseminating content, information, knowledge, or resources to other people through various channels so that more people can review, understand, or benefit from it. The more times a candidate recommendation document is shared, the higher the score of the candidate recommendation document. (5) Number of document transfers: This refers to the number of times a document has been transferred.Forwarding refers to the process of forwarding a document to other people, expanding the reach of the document content and allowing more people to review and understand the document content. The more times a candidate recommendation document is forwarded, the higher the score of the candidate recommendation document. (6) Document Body Quality: Refers to the quality and accuracy of the main content of the document. The quality of the document body can be evaluated and considered from the perspectives of content accuracy, completeness, logic, and clarity of structure, etc. In the present embodiment, a document quality assessment model is used to score the document quality of P candidate recommendation documents, and a score for each candidate recommendation document can be obtained. The document quality assessment model refers to a model or method used to evaluate the quality of a document. For example, the document quality assessment model may include, but is not limited to, a machine learning model, a neural network model, a text classification model, etc. (7) Document Publication Time: Refers to the time when a document is officially posted or published. The later the document publication time of the candidate recommendation document, the higher the score of the candidate recommendation document, i.e., the more recently published the candidate recommendation document, the higher the score, so that the subject can obtain the latest recommendation document when performing a subsequent information search.

[0053] It should be understood that the above recommendation influence factors can be flexibly selected and used alone or in combination. For example, in the above example, (1) and (2) are used in combination. As another example, the above methods (1) and (3) can be used in combination, or the above methods (1), (2), and (5) can be used in combination. For example, the proportion of each recommendation influence factor can be set. For example, the recommendation influence factors include document originality and document text quality, and the proportion of document originality can be set to 30% and the proportion of document text quality can be set to 70%, and the score of the candidate recommended document can be comprehensively determined by setting the weights. By combining multiple recommendation influence factors to sort the candidate recommended documents, it is possible to select better candidate recommended documents as the recommended documents for the target.

[0054] In one implementation, constructing a recommendation recall document library based on the top Y candidate recommendation documents may include performing document topic calculation or content understanding on the top Y candidate recommendation documents to obtain the topic to which each candidate recommendation document belongs, and performing tag calculation or content understanding on the Y candidate recommendation documents to obtain the tag to which each candidate recommendation document belongs, using the topic and / or tag as the tag of each candidate recommendation document, using the topic and / or tag as a search keyword key, and using the candidate recommendation documents (or candidate recommendation document list) that match the corresponding topic and / or tag as index content to construct a recommendation recall document library, where the recommendation recall document library includes at least one of candidate recommendation documents corresponding to different topics and candidate recommendation documents corresponding to different tags, and for example, Y may take a value on the order of one million or tens of millions, for example, Y may take a value on the order of one million or tens of millions.

[0055] 1.3.4) To facilitate online recommendation recall, embodiments of the present application may further establish a mapping relationship between subscription fields and topics, and a mapping relationship between subscription fields and tags. When performing recommendation document recall, the subscription fields can be used as a key to find the mapped topics or tags, and corresponding candidate recommendation documents (or a list of candidate recommendation documents) can be determined according to the topics or tags mapped to the subscription fields. The so-called mapping relationship refers to the ability to find the topics or tags mapped to the subscription fields using the subscription fields. The process of establishing a mapping relationship between subscription fields and topics includes determining the similarity between the subscription fields and topics, and establishing a mapping relationship between the subscription fields and topics if the similarity between the subscription fields and topics is greater than a predetermined similarity. The process of establishing a mapping relationship between subscription fields and tags includes determining the similarity between the subscription fields and tags, and establishing a mapping relationship between the subscription fields and tags if the similarity between the subscription fields and tags is greater than a predetermined similarity. Of course, the mapping relationship between subscription fields and topics and the mapping relationship between subscription fields and titles may also be directly preset. For example, directly set the mapping relationship between the subscription field "Model" and the tag "ChatGPT Model" (Chat Generative Pre-trained Transformer).

[0056] 2) Recommend and customize subscription fields

[0057] In an embodiment of the present application, a subscription field recommendation and customization module is added to the information search interface, which can support the configuration of subscription fields and edit (e.g., delete, modify, etc.) the configured subscription fields, provide one or more candidate search fields for the subject to select, and store the subscription fields configured by the subject, etc. The specific solution of the subscription field recommendation and customization module is as follows:

[0058] When a subject configures a subscription field, the embodiment of the present application provides the following methods to configure the subscription field: (1) Since the subject and the corresponding candidate search fields are already stored in the database, the information search interface includes a search subscription option, and the subject can trigger the search subscription option. When the search subscription option is triggered, one or more candidate search fields corresponding to the subject can be obtained from the database and one or more candidate search fields can be displayed for the subject to select. After the subject selects a candidate search field, the candidate search field can be configured as a subscription field. (2) The subject directly configures a custom subscription field. Specifically, the subject inputs a search field in a search subscription input window, and the input search field can be configured as a subscription field.

[0059] In addition, the subscription field may support editing, which may include modification or deletion. For example, when a subject is no longer interested in a certain subscription field, the subject can delete the subscription field that the subject is not interested in, and can delete the subscription field in response to a delete operation on the subscription field. When a subscription field is deleted, the subscription to the corresponding search field is canceled, indicating that the subject will no longer continuously pay attention to information related to the search field.

[0060] In one implementation, operation records such as configuring selected candidate search fields into subscription fields, configuring custom subscription fields, and deleting subscription fields can be recorded in a subscription field customization database corresponding to the target (e.g., a subscription field customization KV database). Specifically, the target identifier of the target is used as a key, and the operation record corresponding to the target is stored as a value in the subscription field customization database. In another implementation, the target and the subscription field configured by it are saved in a target subscription field database (e.g., a KV database). Specifically, the target identifier of the target is used as a key, and the subscription field configured by the target is stored as a value in the target subscription field database. Subsequently, relevant attention information for the target can be output according to the subscription fields in the target subscription field database.

[0061] It should be understood that any one subject can configure one or more subscription fields according to his / her interests, and thereby continuously obtain relevant information related to these subscription fields.

[0062] 3) Online subscription search process

[0063] 3.1) A subscription field is displayed on the terminal used by the subject, and the subject can trigger (e.g., tap, double tap, etc.) a certain subscription field to enter the information search process. In response to the trigger operation on the subscription field, the server 103 can call a search background CGI (Common Gateway Interface). For example, the search background CGI may be a websearch CGI, and the search background CGI calls the search main dispatch service program mainmixer.

[0064] 3.2) The main dispatch service program mainmixer concurrently invokes a query understanding service program, a target information service program, and a target individual data acquisition service program. Here, the query understanding service program is used to perform query understanding on the subscription field to realize search recall. Here, performing query understanding on the subscription field includes performing query segmentation on the subscription field, building a syntax tree according to the segmentation result, and performing intent recognition on the subscription field to obtain the intent recognition result, thereby completing query understanding. The target information service program is used to acquire target information, which is subsequently used to perform interest-based and need-based sorting on documents retrieved by information retrieval and to search and recall the individual documents. The target individual data acquisition service program is used to acquire individual data, which is used to perform interest-based and need-based sorting on documents retrieved by information retrieval and to search and recall the individual documents.

[0065] 3.3) The main dispatch service program mainmixer can initiate a recall call for documents based on subscription fields within a first preset time period (e.g., the last month or two weeks) to obtain business recall results. The so-called recall call refers to performing information searches based on subscription fields to obtain business recall results. The business recall results can include multiple documents, and the multiple documents can include one or more of search documents and recommended documents. Accordingly, in this embodiment, when initiating a recall call for these two types of documents, two links are involved: one is a search recall link corresponding to the search document, and the other is a recommended recall link corresponding to the recommended document.

[0066] 3.4) The business recall result includes a search recall result, and the method for obtaining the search recall result may include the following steps S11-S15.

[0067] S11: Search Recall: Using the subscription field as the query, search the search document library through the search recall link to retrieve M (M is a positive integer) candidate search documents that match the subscription field within the first preset time period.

[0068] In one implementation, the search document library includes a global search document library, and M candidate search documents whose document publication times are within a first preset time period and match the subscription field are retrieved from the search document library through a search recall link. In another implementation, the search document library includes a global search document library, a new search document library, and a latest search document library. The ob to be used (i.e., the global search document library (gob), the new search document library (fob), or the latest search document library (latest ob)) can be determined according to the first preset time period, and M candidate search documents that match the subscription field can be retrieved from the determined ob. For example, if the first preset time period is the most recent day and the latest search document library is constructed with a daily time period, the ob to be used can be determined as the latest search document library (latest ob). In this case, M candidate search documents that match the subscription field can be retrieved from the latest ob.

[0069] Here, a specific implementation of searching for M candidate search documents that match the subscription field within the first preset time period from the search document library may be to perform fingerprint calculation on the subscription field, obtain the fingerprint features of the subscription field, calculate the similarity between the fingerprint features of the subscription field and the document features of each candidate search document in the search document library, and determine the candidate search documents that correspond to a similarity greater than a similarity threshold as the candidate search documents that match the subscription field.

[0070] In one implementation, an embodiment of the present application may perform a rough sorting of P candidate search documents retrieved from a search document library to obtain M candidate search documents that match the subscription field, where P is greater than or equal to M. Here, rough sorting refers to a process of preliminary screening and sorting a large number of documents (e.g., P candidate search documents) in the search document library to extract a portion of documents (e.g., M candidate search documents) that are relatively highly relevant to the subscription field. For example, searching the search document library for M candidate search documents that match the subscription field within a first preset time period using a search recall link may include: searching the search document library for P candidate search documents whose document publication times are within the first preset time period and that match the subscription field using the search recall link; sorting the P candidate search documents in descending order of their relevance to the subscription field; and determining M candidate search documents that match the subscription field based on the sorted P candidate search documents. Specifically, the top M candidate search documents can be selected from the sorted P candidate search documents as the M candidate search documents that match the subscription field.

[0071] S12: According to the target browsing history record, perform a filtering process on M candidate search documents to obtain L unviewed candidate search documents. One implementation is to call a viewed document filter constructed and obtained based on the browsing history record, and perform a filtering process on M candidate search documents to obtain L unviewed candidate search documents. Another implementation is to perform a fingerprint calculation on M candidate search documents, obtain fingerprint features of each of the M candidate search documents, and compare them with the document features of each viewed document in the viewed document database, and perform a filtering process on candidate search documents that match the fingerprint features of the viewed documents among the M candidate search documents to obtain L unviewed candidate search documents, where L is a positive integer and is less than or equal to M.

[0072] S13: Sorting the L unviewed candidate search documents to obtain L sorted candidate search documents, where sorting here can be understood as performing precise sorting (i.e., precise sorting of search) according to target information, and precise sorting (i.e., precise sorting of search) refers to the process of sorting search results according to a certain standard (e.g., target information). In one implementation, sorting is performed on the L unviewed candidate search documents according to the target information, and if the target information indicates that cooking is of high interest (the higher the attention, the greater the interest in the subject), then the candidate documents related to cooking can be sorted higher among the L unviewed candidate search documents, and other candidate search documents can be sorted lower among the L unviewed candidate search documents.

[0073] S14: Perform a recall process of the individual searched document based on the individual data of the target to obtain the individual searched document. The individual data includes, but is not limited to, like documents, shared documents, forwarded documents, other targets having a target-target relationship, and documents focusing on other targets. Performing a recall process of the individual searched document based on the individual data of the target to obtain the individual searched document may include obtaining a document according to the individual data and determining the obtained document as the individual searched document.

[0074] For example, the individual data may include other objects with which the object relationship exists, and the object relationship may include, but is not limited to, a friend relationship in a social application, a featured relationship, etc. Performing a recall process for the individual searched document based on the individual data of the object and obtaining the individual searched document may include obtaining documents viewed by other objects and setting the obtained documents as the individual searched document. As another example, the individual data may include an account of interest, and performing a recall process for the individual searched document based on the individual data of the object and obtaining the individual searched document may include obtaining documents related to the account of interest and setting the obtained documents related to the account of interest as the individual searched document.

[0075] S15: Perform a merge sort process on the sorted L candidate search documents and the individual search documents to obtain a search recall result, where the search recall result may include one or more search documents.

[0076] In one implementation, the fusion sorting process includes at least one of the following: (1) performing a duplicate elimination process on the sorted L candidate search documents and the individual search documents to obtain the duplicate elimination documents, and then performing a sort process on the duplicate elimination documents to obtain the search recall results; (2) adjusting the order of the sorted L candidate search documents based on the individual search documents to obtain the search recall results. For example, the individual search documents include document 1 and document 2, and the L candidate search documents include document 1 and document 3. The order of document 1 in the sorted L candidate search documents can be adjusted according to document 1 in the individual documents (for example, adjusting document 1 to be positioned higher in the L candidate search documents), thereby obtaining the search recall results.

[0077] 3.5) For a recommendation recall link, the task recall result may further include a recommendation recall result, and the method for obtaining the task recall result may include the following steps S21-S24.

[0078] S21: Recommendation recall: Using the subscription field as an index keyword (key), determine the topic or tag mapped to the subscription field in the recommendation recall document library provided by the recommendation recall link, and search for candidate recommendation documents that match the subscription field based on the mapped topic or tag in the recommendation recall document library provided by the recommendation recall link, thereby realizing document recall from multiple recall directions such as topic and tag. For example, the subscription field is "Model", and determine the tag "ChatGPT Model" mapped to the subscription field "Model" in the recommendation recall document library provided by the recommendation recall link. Search for candidate recommendation documents that belong to the tag "ChatGPT Model" based on the tag "ChatGPT Model" in the recommendation recall document library, and determine the found candidate recommendation documents that belong to the tag "ChatGPT Model" as candidate recommendation documents that match the subscription field.

[0079] S22: Roughly sorting recommendations based on the topic or tag mapped to the subscription field or the candidate recommendation subscription information. One implementation method involves obtaining sorting parameters and sorting the retrieved matching candidate recommendation documents according to the sorting parameters to obtain the sorted candidate recommendation documents. Here, the sorting parameters include text quality and compatibility with the subscription field. Taking compatibility with the subscription field as an example, sorting the retrieved matching candidate recommendation documents according to the sorting parameters to obtain the sorted candidate recommendation documents includes sorting the retrieved matching candidate recommendation documents in descending order of compatibility with the subscription field to obtain the sorted candidate recommendation documents. For example, the number of retrieved matching candidate recommendation documents is three, Document 1, Document 2, and Document 3. Assume that Document 1's compatibility with the subscription field is greater than Document 2's compatibility with the subscription field, and Document 2's compatibility with the subscription field is greater than Document 3's compatibility with the subscription field. The sorted candidate recommendation documents are Document 1, Document 2, and Document 3, in that order.

[0080] S23: Filter the sorted candidate recommendation documents based on the browsing history record to obtain unviewed candidate recommendation documents. Here, the specific implementation method of step S23 is similar to the specific implementation method of step S12, so detailed description is omitted here.

[0081] S24: Perform precision sorting of recommendations on unviewed candidate recommendation documents. As one implementation, perform re-sorting of unviewed candidate recommendation documents according to the target information and individual data to obtain a recommendation recall result. The unviewed candidate recommendation documents after the sorting process can be the recommended documents, and the recommendation recall result may include one or more recommendation documents. Specifically, perform weight calculation on unviewed candidate recommendation documents according to the target information and individual data to obtain the weights of the unviewed recommendation documents, and perform re-sorting of unviewed candidate recommendation documents in descending order of weight to obtain the recommendation recall result. As one implementation, performing weight calculation on unviewed candidate recommendation documents according to the target information and individual data may determine unviewed candidate recommendation documents that the target is interested in according to the target information and individual data, and determine the weight of the interested unviewed candidate recommendation document as a relatively large value, while setting the weight of the other unviewed candidate recommendation documents as a relatively small value.

[0082] It should be understood that the browsing history includes the viewed documents. The documents may be videos, texts, audios, etc. Therefore, in the embodiment of the present application, if the documents are texts and videos, for example, the search recall links and recommendation recall links corresponding to the texts can be increased, and the search recall links and recommendation recall links corresponding to the videos can be increased, thereby realizing the business recall results obtained by performing information searches from two directions, namely, videos and texts, based on the subscription fields. The business recall results obtained by performing information searches from two directions, namely, videos and texts, based on the subscription fields can refer to the business recall results obtained by performing information searches based on the subscription fields, and there is no limitation on this in the embodiment of the present application.

[0083] In addition, in the present embodiment, the viewed documents can be filtered as in steps S12 and S23. Of course, in the present embodiment, the viewed documents can be directly recalled without filtering and folded when output. In this case, the search recall link searches the search document library for M candidate search documents that match the subscription field within the first preset time period through the search recall link, and then directly sorts the M candidate search documents to obtain the sorted M candidate search documents. The individual search documents are then recalled based on the target individual data to obtain the individual search documents. The sorted M candidate search documents and the individual search documents are then merged and sorted to obtain the search recall results. In the recommendation recall link, the sorted candidate recommendation documents are re-sorted based on the target information and individual data to obtain the recommendation recall results. In addition, in the embodiments of the present application, documents that are not of interest to the subject can be determined according to the subject's behavior log, for example, the behavior log includes marking documents corresponding to a certain topic or tag as uninteresting documents (for example, documents whose attention level is lower than an attention level threshold), in which case, filtering can be performed on this type of document as well. At this time, the processing of uninteresting documents can be realized by referring to the processing process of viewed documents, which will not be discussed again here.

[0084] In steps 3.4) and 3.5), the viewed documents or documents of no interest are filtered or collapsed, and the latest content related to the target information or interest needs of the target is actively displayed, so as to satisfy the continuous interest consumption needs under the target's subscription field and to avoid the target from repeatedly viewing the viewed documents as much as possible.

[0085] It should be understood that in the embodiments of the present application, the recommendation recall link and the search recall link may be two independent links. Naturally, the recommendation recall link and the search recall link are the same search link. In this case, the document topic or tag can be used as the document text field in the search recall link, and the document text field can be vectorized to obtain a vector corresponding to the text field. A search index can then be constructed based on the vector corresponding to the document text field to obtain a recommendation recall document library. When performing information retrieval, the search recall link can be used to directly retrieve documents from the recommendation recall document library and the search document library. Document recall can be completed using multiple search recall methods.

[0086] 3.6) Subscription recommendation fusion processing: After obtaining the search recall results and the recommendation recall results, a fusion process is performed on the search recall results and the recommendation recall results to obtain business recall results, and the business recall results are returned to mainmixer. In one implementation, performing a fusion process on the search recall results and the recommendation recall results may include directly determining the search recall results and the recommendation recall results as business recall results. In another implementation, performing a fusion process on the search recall results and the recommendation recall results may include performing a duplicate elimination process on the search documents in the search recall results and the recommended documents in the recommendation recall results to obtain business recall results. In yet another implementation, performing a fusion process on the search recall results and the recommendation recall results may include performing a fusion sort on the search documents in the search recall results and the recommended documents in the recommendation recall results according to a fusion strategy to obtain business recall results. The fusion strategy includes an attention fusion display rule, a subscription field relevance fusion rule, and a document diversity fusion display rule.

[0087] When the fusion display rules include attention fusion display rules, performing fusion sorting on the retrieved documents in the search recall results and the recommended documents in the recommendation recall results according to the fusion strategy to obtain business recall results includes performing fusion sorting on the retrieved documents in the search recall results and the recommended documents in the recommendation recall results in descending order of attention to obtain business recall results. For example, if the attention of the retrieved documents in the search recall results is greater than that of the recommended documents in the recommendation recall results, the retrieved documents are sorted before the recommended documents in descending order of attention to obtain business recall results. Here, the attention can be determined according to the real-time interest needs of the subject. For example, if the real-time interest needs indicate that the subject's attention direction is cooking, in this case the retrieved documents are documents related to cooking, and in this case the attention of the retrieved documents is greater than that of the recommended documents.

[0088] When the fusion display rule includes a subscription field relevance fusion rule, the subscription field relevance fusion rule may include the relevance of the subscription field to the topic or the relevance of the subscription field to the tag. For example, if the subscription field relevance fusion rule includes the relevance of the subscription field to the topic, performing fusion sorting on the searched documents in the search recall results and the recommended documents in the recommendation recall results according to the fusion strategy to obtain business recall results includes determining the relevance of the searched documents to the topic of the subscription field, and determining the relevance of the recommended documents to the topic of the subscription field, and performing fusion sorting on the searched documents and the recommended documents in order of relevance to the topic of the subscription field, to obtain business recall results.

[0089] When the fusion display rules include document diversity fusion display rules, for example, the document diversity fusion display rules include that document diversity includes the distribution of the number of documents from the same account, and performing fusion sorting on the retrieved documents in the search recall results and the recommended documents in the recommendation recall results according to the fusion strategy to obtain business recall results includes performing fusion processing on the retrieved documents in the search recall results and the recommended documents in the recommendation recall results according to the distribution of the number of documents from the same account to obtain business recall results. For example, suppose the number of searched documents in the search recall result is four, where three of the searched documents belong to the same account, and the number of recommended documents in the recommendation recall result is two, and these two recommended documents belong to the same account, and the document count distribution of the same account is two or less. In this case, a fusion process can be performed on the four searched documents and two recommended documents according to the document count distribution of the same account. The fusion process here may include deleting one of the three searched documents that belong to the same account, so that the final business recall result includes three searched documents and two recommended documents.

[0090] 3.7) After mainmixer obtains the task recall result, the task recall result can be directly output. Alternatively, after mainmixer obtains the task recall result, it performs mixed sorting of multiple task results on the documents in the task recall result, obtains the mixed sorted task recall result, and outputs the mixed sorted task recall result. As an implementation method, mixed sorting of multiple task results can be performed on the documents in the task recall result based on real-time interest needs and / or document diversity needs. For example, mixed sorting of multiple task results can be performed on the documents in the task recall result based on document diversity needs. For example, the documents in the task recall result include text and video, and the document diversity needs include alternating videos and text. In this case, the text and video in the task recall result can be alternated based on the document diversity needs, and the mixed sorted task recall result can be obtained.

[0091] 3.8) Optionally, if special style customization is set in the subscription search, mainmixer calls the style customization service program (mergesvr) to perform style customization on the task recall results after mixed sorting, and feeds back the customized task recall results to the terminal, causing the terminal to output the customized task recall results. Style customization includes, but is not limited to, topic aggregation style, tag aggregation style, etc. For example, if the style customization includes a topic aggregation style, topic aggregation can be performed on documents in the task recall results, and documents belonging to the same topic can be displayed in one area.

[0092] 4) Subscription search results feedback

[0093] After obtaining the business recall results, the subject can call a feedback CGI (webquery CGI) during the process of viewing the business recall results to provide feedback on the viewed documents. For example, the subject can mark documents that are of no interest or interest, respond to the marking operation, and record the subject's feedback in the database (in the KV database). In the subscription search process 3.6), during the process of performing fusion sorting on the search recall results and the recommendation recall results, the subject's feedback can be adjusted according to the subject's feedback during the process of performing fusion sorting on the search recall results and the recommendation documents in the recommendation recall results, or can be directly used to filter the search documents in the search recall results and the recommendation documents in the recommendation recall results. If the subject's feedback indicates that the document is not of interest, the weight adjustment here may be to reduce the weight of the searched document or the recommended document when the similarity between the searched document and the marked document is relatively high; if the subject's feedback indicates that the document is of interest, the weight adjustment here may be to increase the weight of the searched document or the recommended document when the similarity between the searched document and the marked document is relatively high.

[0094] 5) Content update notification

[0095] When there is an update of the content (document) related to a subscription field, a content update notification for the subscription field can be generated and output, and the content update notification can notify a target that there is a content update in the subscription field that the target is paying attention to, thereby facilitating the target's continuous consumption (for example, facilitating the target's browsing and inquiry, etc.). Referring to Figure 2b, a flow diagram of a subscription update notification provided by an exemplary embodiment of the present application, the entire process of the subscription update notification is as follows:

[0096] 5.1) Subscription notification dispatch service program: The server's subscription notification dispatch service program scans the target subscription field database at minute intervals, performs time-distributed dispatching using the target as a dimension, and adds the notifications to the dispatch queue. That is, different targets are queued in a uniformly distributed manner over time, avoiding system pressure due to update calculations. Distributed dispatching refers to a dispatch strategy that disrupts the original dispatch order through randomization or other methods to improve system performance or fairness. For example, subscription fields corresponding to 100 targets can be added to the dispatch queue every 1 minute or every 2 minutes.

[0097] 5.2) Subscription Search: The server obtains the target subscription field from the dispatch queue and can re-recall documents using the search recall link and the recommendation recall link. In one implementation, the search recall link is used to search for one or more local search documents that match the subscription field from the local search document library, and a first filtering result is obtained by filtering the one or more local search documents retrieved through the search recall link based on the target browsing history. Because document updates occur relatively quickly, there may be documents that have been updated and notified previously. In this situation, a filtering process is performed on the one or more local search documents retrieved through the search recall link based on the browsing history and the notified documents, i.e., the notified documents are filtered out from the one or more local search documents, thereby obtaining a first filtering result.

[0098] Here, the process of searching for one or more local search documents that match the subscription field from the local search document library is similar to the process of searching for M candidate search documents that match the subscription field from the search document library, so a detailed description will be omitted here.

[0099] 5.3) Determine the topic or tag mapped to the subscription field from the recommendation recall document library provided by the recommendation recall link, and search for one or more candidate recommendation documents that match the subscription field in the recommendation recall document library based on the mapped topic or tag, and perform a filtering process on one or more candidate recommendation documents searched by the recommendation recall link based on the browsing history to obtain a second filtering result. The filtering process on one or more candidate recommendation documents searched by the recommendation recall link here is similar to 5.2), so it will not be described again.

[0100] 5.4) Generating a content update notification: The first filtering result and the second filtering result are merged to obtain an update recall result. If the update recall result is not an empty set, a content update notification for the subscription field is generated based on the update recall result. In one implementation, because updates occur relatively quickly, the previously obtained update recall result may not yet have generated a content update notification. In this case, the previous update recall result and the current update recall result are merged to obtain a content update notification. That is, the content update notification includes the update recall result for which no timely update notification was previously received and the current update recall result. In another implementation, a content update notification for the subscription field is directly generated based on the update recall result. The content update notification may include one or both of the number of updated documents and the recommendation reason. When the content update notification includes the recommendation reason, a content tag clustering process can be performed on the documents in the update recall result to obtain the type to which the document in the update recall result belongs. A recommendation reason is generated based on the type to which the document belongs. For example, the documents in the update recall results are clustered to obtain the type related to the technical principles of the model to which the documents in the update recall results belong. In this case, the existence of a document update related to the technical principles of the model can be output as the reason for recommendation.

[0101] 5.5) The server calls a push service program to output a content update notification for the subscription field. For example, the content update notification may be output in a highlighting manner (e.g., a colored dot).

[0102] By actively outputting content update notifications, the subject does not need to actively check whether there are content updates in the subscription field configured in the information search interface, which simplifies information search operations, improves information search efficiency, and better enhances the subject's experience.

[0103] Next, a related discussion will be given on the information processing method provided by the embodiment of the present application.

[0104] 3a, there is shown a flow chart of an information processing method provided by an exemplary embodiment of the present application. The information processing method may be executed by a computer device in the information processing system, and the computer device may be a terminal in the information processing system. The information processing method may include the following steps S301-S303:

[0105] S301: Display an information search interface, which is used to perform information search based on a search field.

[0106] Here, the search field may include keywords, complete sentences, etc. For example, the search field may be Xiaoming, music, cooking, etc., and the search field may be "How long is the food festival?" For example, see FIG. 3b, which is a schematic diagram of an embodiment of the present application providing featured information related to subscription information. As shown in FIG. 3b, the information search interface provides a search field input control (component) 31 in the information search interface 301, and a subject can input a search field in the search field input control 31 to perform an information search based on the search field.

[0107] S302: Display a subscription field, which refers to a configured search field that requires continuous attention.

[0108] In one implementation, the target can configure subscription fields of interest in an information search interface, and the terminal can correspondingly obtain the configured subscription fields and display the subscription fields. Here, displaying the subscription fields may include at least one of the following: (1) Displaying the subscription fields in the information search interface, and displaying the subscription field 32 in the information search interface, as shown in FIG. 3b. The display position of the subscription field 32 can be flexibly set. For example, the subscription field 32 may be displayed at a fixed position in the information search interface, or the subscription field 32 may be displayed floating in the information search interface. This embodiment of the present application is not limited to this. (2) Displaying the subscription field in a first interface independent of the information search interface. The first interface may be any interface other than an information search interface, for example, the information search interface is displayed on the client terminal, and the first interface may be a main interface of the client terminal independent of the information search interface, a service interface, a conversation message interface, etc., and the display position of the subscription field can be flexibly set, for example, the subscription field may be displayed at a fixed position on the first interface, or the subscription field may be displayed floating on the first interface, and in the embodiments of the present application, there is no limitation on the display interface of the subscription field and the display position of the subscription field on the display interface.

[0109] In an embodiment of the present application, the number of subscription fields may be N, where N is a positive integer. Displaying the subscription fields may include at least one of the following: (1) Tiling N subscription fields. For example, FIG. 4a is a schematic diagram illustrating N subscription fields provided by an exemplary embodiment of the present application. In FIG. 4a, assuming that the subscription fields are displayed in an information search interface, subscription field 1, subscription field 2, and subscription field 3 are tiled in area 33 of information search interface 301. (2) Arranging the N subscription fields in order of the earliest configuration time. For example, assuming that N=3, subscription field A was configured earlier than subscription field B, which is configured earlier than subscription field C, and the subscription fields are displayed in an information search interface, subscription field A, subscription field B, and subscription field C can be arranged in order of the earliest configuration time in the information search interface. (3) Arranging the N subscription fields in order of the most popular order. Here, the higher the attention level of a subscription field, the greater the interest of the target in the subscription field. For example, the attention level of subscription field A is higher than the attention level of subscription field B, which is higher than the attention level of subscription field C. FIG. 4b is a schematic diagram showing N subscription fields provided by another exemplary embodiment of the present application.For example, when subscription fields are displayed in an information search interface, N=3, and subscription field A, subscription field B, and subscription field C are displayed in the information search interface in descending order of popularity, as shown in FIG. 4b. (4) N subscription fields are displayed randomly. (5) N subscription fields are classified and displayed according to the topics to which they respectively belong. As an implementation method, when subscription fields are displayed in an information search interface, classifying and displaying N subscription fields according to the topics to which they respectively belong may include displaying subscription fields belonging to different topics in different areas of the information search interface, and displaying subscription fields belonging to the same topic in the same area. For example, when subscription fields are displayed in an information search interface, as shown in FIG. 4c, a schematic diagram of displaying N subscription fields provided by another exemplary embodiment of the present application is provided. In Figure 4c, subscription field a and subscription field b belong to the same topic, and subscription field c and subscription field d belong to the same topic, and in the information search interface 302, subscription field a and subscription field b are displayed in area 34, and subscription field c and subscription field d are displayed in area 35. (6) Displaying N subscription fields in a collapsed state. Taking the display of subscription fields in an information search interface as an example, Figure 4d is a schematic diagram of N subscription fields in a collapsed state provided by one exemplary embodiment of the present application.In FIG. 4d, the information search interface can display two (N=2) subscription fields in a collapsed state, and the collapsed subscription fields can be expanded and displayed when triggered. The information search interface in FIG. 4d includes a subscription field expansion option 36, which displays two collapsed subscription fields (i.e., subscription field 1 and subscription field 2) in response to a trigger operation on the subscription field expansion option 36. (7) Displaying the N subscription fields in descending order of field length. For example, if the field length of the subscription field "XX City" is 4 and the field length of the subscription field "XXXX Cuisine" is 6, and the subscription fields are displayed in the information search interface, the subscription field "XXXX Cuisine" and the subscription field "XX City" will be displayed in descending order of field length in the information search interface.

[0110] S303: Output noteworthy information related to the subscription field.

[0111] In one implementation, the number of subscription fields is N, and outputting the attention information related to the subscription fields includes, in response to a selection operation on the subscription fields, outputting the attention information related to the subscription fields selected by the selection operation. Here, the attention information includes, but is not limited to, task recall results obtained by performing an information search based on the subscription fields and content update notices. The task recall results may include multiple documents that match the subscription fields, and the multiple documents may include recommended documents and / or searched documents. For example, in FIG. 3b, document 321, document 322, document 323, and document 324 related to the subscription fields are output. The content update notice can be used to prompt that there is an update to the document related to the subscription field, and the content update notice may include one or two of the number of updated documents and a recommendation reason.

[0112] In an embodiment of the present application, outputting the featured information related to the subscription field may include at least one of the following: (1) outputting the featured information in a fixed area of ​​the information search interface. The fixed area may be any area of ​​the information search interface, such as the upper area, middle area, or right area, and the embodiment of the present application is not limited thereto. For example, the featured information includes a task recall result, and the task recall result includes document 1 and document 2 that match the subscription field. FIG. 5a is a schematic diagram of an exemplary embodiment of the present application displaying multiple documents that match the subscription field in a fixed area. In FIG. 5a, document 1 and document 2 that match the subscription field are displayed in a fixed area 51 of the information search interface 501. (2) outputting the featured information in a floating area of ​​the information search interface. For example, the featured information includes a content update notification. FIG. 5b is a schematic diagram of an exemplary embodiment of the present application displaying a content update notification in a floating area. In FIG. 5b, a content update notification 1 related to the subscription field is displayed in a floating area 52 of an information search interface 501. (3) Attention information is output in a second interface independent of the information search interface. For example, attention information includes task recall results, and the task recall results include documents 1 and 2 that match the subscription field. As shown in FIG. 3b, documents 321, 322, 323, and 324 that match the subscription field are displayed in a second interface 302 independent of the information search interface 301. The second interface may be any interface other than the information search interface. For example, the information search interface is displayed on a client terminal, and the second interface may be a main interface, a service interface, a conversation message interface, etc. of the client terminal that are independent of the information search interface. In the embodiments of the present application, there is no limitation on the second interface.

[0113] In an embodiment of the present application, an information search interface is displayed, which is used to perform information search based on search fields, and displays subscription fields, where the subscription fields refer to configured search fields, and outputs attention information related to the subscription fields. This method not only enables information search based on search fields in the information search interface, but also allows the subject to configure the search fields as subscription fields according to their needs for continuous attention. By displaying the subscription fields, the configured search fields that need to be continuously paid attention can be presented, and the subject can know the search fields that are continuously interested. Furthermore, by continuously performing information search through the subscription fields, the subject can obtain and output the information that is continuously interested in, which enriches the information search method and at the same time improves the flexibility of information search, thereby meeting the subject's needs for continuous attention to information.

[0114] Next, a detailed implementation process for obtaining the subscription field provided by the embodiment of the present application will be described.

[0115] In one embodiment, obtaining the subscription field may include, but is not limited to, the following manners:

[0116] (1) In an information search interface, a search subscription option is provided, and a target can trigger the search subscription option to perform a configuration operation and enter a subscription field in a search subscription input window. In this implementation, obtaining a subscription field includes displaying a search subscription input window in response to the search subscription option being triggered, then receiving a first search field entered in the search subscription input window, and configuring the first search field as a subscription field. Here, the triggering manner of the search subscription option may be tapping, double tapping, voice triggering, swiping, etc. For example, see FIG. 6A, which is a schematic diagram of configuring a subscription field provided by an exemplary embodiment of the present application. In FIG. 6a, a search subscription option 61 is provided in the information search interface, and the subject taps the search subscription option 61. In response to the search subscription option 61 being triggered, a search subscription input window 601 is displayed, and the subject can input a first search field "piano music" in the search subscription input window 601. Correspondingly, the first search field "piano music" input in the search subscription input window 601 is received, and the first search field "piano music" is configured as a subscription field.

[0117] (2) In the information search interface, a search subscription option is provided, and acquiring subscription fields includes displaying one or more candidate search fields in response to the search subscription option being triggered, receiving a selection operation on the candidate search fields, and configuring the selected candidate search fields as subscription fields. Here, the one or more candidate search fields can be directly displayed in the information search interface, or the one or more candidate search fields can be displayed in a search subscription input window, or the one or more candidate search fields can be displayed in a search subscription interface independent of the information search interface, and there is no limitation thereon in the embodiment of the present application.

[0118] 6b is a schematic diagram illustrating how to set a subscription field provided in another exemplary embodiment of the present application. In FIG. 6b, a search subscription option 61 is provided in an information search interface. A subject taps the search subscription option 61. In response to the search subscription option 61 being triggered, two candidate search fields are displayed, which are candidate search field 1 and candidate search field 2, respectively. The subject can select one or both of the two candidate search fields to configure as subscription fields. Assume that the subject selects candidate search field 1, and a selection operation for candidate search field 1 is received, and the selected candidate search field 1 is configured as a subscription field.

[0119] Here, the method of obtaining one or more candidate search fields includes obtaining one or more candidate search fields corresponding to the target identifier from a database based on the target identifier, and the candidate search fields are obtained by collecting behavior logs of the target within a certain time period (e.g., day level, hourly level, etc.) and calculating candidate search subscription fields for the behavior logs using content tag clustering technology.

[0120] It should be understood that when the subject triggers the search subscription option, one or more candidate search subscription fields may be displayed in the search subscription input window, and in this situation, the subject may custom input the first search field that needs to be configured as a subscription field according to their interest needs in the search subscription input window, or may directly select the candidate search field that needs to be configured as a subscription field from the one or more candidate search fields.

[0121] (3) In the information search interface, a search field input control is included, and in this implementation method, obtaining the subscription field includes receiving a second search field input in the search field input control, responding to a configuration operation on the second search field, and configuring the second search field as a subscription field.

[0122] Here, the configuration operation for the second search field may include any one of the following: A: A subscription setting option is displayed in the information search interface, and the configuration operation for the second search field may include a trigger operation for the subscription setting option. Here, the subscription setting option can be displayed in the information search interface after receiving the second search field input in the search field input control. Or, the subscription setting option is always displayed in the information search interface and is used to configure the second search field input in the search field input control as a subscription field. For example, refer to FIG. 6c, which shows a schematic diagram of setting a subscription field provided by another exemplary embodiment of the present application. In FIG. 6c, the information search interface includes a search field input control 62, and a subject can input the second search field "artificial intelligence model" in the search field input control 62. Correspondingly, the second search field "artificial intelligence model" input in the search field input control is received, and a subscription setting option 63 is displayed in the information search interface, and the second search field "artificial intelligence model" is configured as a subscription field in response to the trigger operation of the displayed subscription setting option 63. B: Performing operations such as tapping and double-tapping on a preset area of ​​the information search interface, where the preset area may be the middle area, left area, right area, etc. of the information search interface, and there is no limitation thereon in the embodiment of the present application. C: Performing operations such as double-tapping, moving, and tapping on the second search field. For example, see FIG. 6d, which is a schematic diagram of setting a subscription field provided by another exemplary embodiment of the present application.6d, the information search interface includes a search field input control 62, and the subject can input a second search field "artificial intelligence model" in the search field input control 62. Correspondingly, the second search field "artificial intelligence model" input in the search field input control is received, and the subject taps the second search field "artificial intelligence model" and configures the second search field "artificial intelligence model" as a subscription field in response to the tap operation on the second search field. D: Input operation of a specific gesture in the information search interface, and the specific gesture may include, but is not limited to, an M gesture, an OK gesture, etc.

[0123] (4) After the target performs an information search based on the search field, the target will record the history search field. Therefore, in an embodiment of the present application, obtaining a subscription field may include displaying at least one history search field of the information, where the history search field refers to a search field used in the history information search process, and receiving a selection operation for the history search field and configuring the selected history search field as a subscription field. For example, see FIG. 6e, which is a schematic diagram of setting subscription fields provided by an exemplary embodiment of the present application. In FIG. 6e, two history search fields are displayed in the information search interface, namely, the history search field "Model" and the history search field "Art and Design", respectively. The target can select one or both of the history search field "Model" and the history search field "Art and Design" to set them as subscription fields. If the target selects the history search field "Model", a selection operation for the history search field "Model" is correspondingly received, and the history search field "Model" is set as a subscription field.

[0124] In an embodiment of the present application, the subscription field may support editing, where editing may include modification or deletion. In this case, when editing includes modification, in one implementation, a modification option for the subscription field is displayed, and the target can trigger the modification option by tapping, double-tapping, etc. In response to the trigger operation for the modification option, the subscription field can be set to an editable state, and the subscription field is modified when a modification operation for the subscription field is received.

[0125] When the editing includes deletion, in one implementation, a delete option for the subscription field can be displayed, and the subject can trigger the delete option by tapping, double-tapping, etc., and the subscription field is deleted in response to the trigger operation on the delete option. For example, FIG. 7a is a schematic diagram of deleting a word in a subscription field provided by an exemplary embodiment of the present application. In FIG. 7a, subscription field 1 and subscription field 2 are displayed in the information search interface, and at the same time, a delete option 71 corresponding to subscription field 1 and a delete option 72 corresponding to subscription field 2 are also displayed. When the subject taps the delete option 72 corresponding to subscription field 2, the subscription field 2 is deleted in the information search interface in response to the tap operation on the delete option 72 corresponding to subscription field 2. In another implementation, the subject can move the subscription field that needs to be deleted to a fixed area, and the moved subscription field can be deleted from the information search interface in response to the move operation on the subscription field.

[0126] In some possible embodiments, when a subscription field is changed by editing, the changed subscription field needs to be refreshed and displayed. For example, in FIG. 7a, after subscription field 2 is deleted, the changed subscription fields are refreshed and displayed in the information search interface, i.e., only field 1 is displayed in the information search interface. The number of changed subscription fields is M, where M is less than or equal to N. Refreshing and displaying the changed subscription fields may include at least one of the following: (1) tiling the M changed subscription fields; (2) arranging and displaying the M changed subscription fields in descending order of the time when the M changed subscription fields were configured; (3) arranging and displaying the M changed subscription fields in descending order of attention; (4) randomly displaying the M changed subscription fields; or (5) arranging and displaying the M changed subscription fields in descending order of the field lengths of the M changed subscription fields.

[0127] The above describes an example in which the subscription field is displayed in an information search interface, and it should be understood that when the subscription field is changed by editing, the changed subscription field can also be refreshed and displayed in a second interface independent of the information search interface.

[0128] As described above, in the embodiment of the present application, the subscription field setting method is very flexible. For example, when setting the subscription field, it is supported that the subject sets the subscription field in a custom manner and selects a candidate search field to configure it as the subscription field. As another example, when setting the subscription field, it is supported that the first search field entered in the search subscription input window is configured as the subscription field. Flexible configuration of the subscription field can promote the consumption of content in situations where the subject has continuous interest needs.

[0129] Next, a related discussion will be given on outputting attention information related to subscription fields, to which the embodiments of the present application relate.

[0130] As mentioned above, the attention information includes the business recall result obtained by performing an information search based on the subscription field, and the content update notice, etc. In one possible embodiment, the attention information includes the business recall result obtained by performing an information search based on the subscription field. The business recall result may include a plurality of documents that match the subscription field, and the documents may include text, web pages, videos, images, audio, etc. In this embodiment, outputting the attention information related to the subscription field includes at least one of the following:

[0131] (1) Arrange and display multiple documents in a random order.

[0132] (2) Multiple documents are displayed in descending order of relevance between each document and the subscription field. In one implementation, the relevance between each document and the subscription field can be determined, and then the multiple documents are displayed in descending order of relevance between each document and the subscription field. For example, FIG. 7b is a schematic diagram showing the display of multiple documents that match the subscription provided by an exemplary embodiment of the present application. In FIG. 7b, the business recall result includes three documents, where document 1 has a relevance of 98% between the subscription field, document 2 has a relevance of 88%, and document 3 has a relevance of 80% between the subscription field and the document. Document 1, document 2, and document 3 are displayed in descending order of relevance between each document and the subscription field.

[0133] (3) Multiple documents are displayed in descending order of attention. Attention can be understood as the degree of interest of the subject, and the higher the subject's interest, the higher the attention. For example, the subscription field is "ChatGPT model" (Chat Generative Pre-trained Transformer), and multiple documents matching the subscription field include Document 1 "Technical Implementation of the ChatGPT Model" and Document 2 "AI Model Overview." The subject is interested in the ChatGPT model, and the attention of Document 1 "Technical Implementation of the ChatGPT Model" is higher than that of Document 2 "AI Model Overview." Therefore, Document 1 "Technical Implementation of the ChatGPT Model" and Document 2 "AI Model Overview" are displayed in descending order of attention.

[0134] Here, the document's attention level can be determined according to the target's real-time interest needs. The real-time interest needs can be obtained by performing content tag clustering processing according to the target's behavior log and target information within the time period closest to the target's system time (e.g., the last three days, the last week). For example, if the real-time interest needs indicate that the target is interested in the ChatGPT model, a relatively high attention level can be set for documents related to the ChatGPT model.

[0135] (4) Displaying multiple documents side by side according to the need for document diversity. Here, document diversity may include, but is not limited to, document carrier diversity and document source diversity. Document carriers may include, for example, video, text, web pages, and audio. Document sources may include accounts to which documents belong and document posting platforms. For example, document diversity needs may include document carrier diversity needs, i.e., document carriers may be required to include alternating text and video. FIG. 7c illustrates a schematic diagram of multiple documents displayed side by side according to an exemplary embodiment of the present application. In FIG. 7c, the multiple documents include document A, document B, and document C. The document carrier for document A is a video, the document carrier for document B is a video, and the document carrier for document C is a text. According to the need for document carrier diversity, document A, document C, and document B are displayed side by side. For example, the need for document diversity includes the need for diversity of accounts to which documents belong, i.e., requiring that no more than two documents belonging to the same account be displayed at one time, and the multiple documents include document A, document B, document C, and document D, where document A, document B, and document C all belong to account A, and document D belongs to account B. According to the need for diversity of accounts to which documents belong, document A, document B, document D, and document C are displayed side by side.

[0136] (5) Clustering and displaying multiple documents according to the topics or tags to which the documents belong. In one implementation, clustering and displaying multiple documents according to the topics to which the documents belong includes classifying and displaying documents belonging to different topics in different regions, and clustering and displaying documents belonging to the same topic in the same region. For example, documents 1 and 2 belong to topic 1, and documents 3 and 4 belong to topic 2. FIG. 7d is a schematic diagram of clustering and displaying multiple documents provided by one exemplary embodiment of the present application. In FIG. 7d, documents 1 and 2 are clustered and displayed in region 73 according to the topics to which the documents belong, and documents 3 and 4 are clustered and displayed in region 74. In another implementation, clustering and displaying multiple documents according to the tags to which the documents belong includes classifying and displaying documents belonging to different tags in different regions, and clustering and displaying documents belonging to the same tag in the same region. For example, if the tag to which documents 1 and 2 belong is "beautiful scenery" and the tag to which documents 3 and 4 belong is "tree planting," documents 1 and 2 are clustered and displayed in a first area, and documents 3 and 4 are clustered and displayed in a second area.

[0137] In some possible implementations, the plurality of documents may include viewed documents and unviewed documents, and outputting attention information related to the subscription field includes at least one of the following:

[0138] (1) Filtering viewed documents and displaying only unviewed documents. One implementation method is to, after obtaining the task recall results, filter the viewed documents from the task recall results according to the browsing history record to obtain unviewed documents and display only the unviewed documents on the interface. Specifically, the browsing history record includes fingerprint features of the viewed documents. Filtering the viewed documents from the task recall results according to the browsing history record to obtain unviewed documents involves determining the fingerprint features of the documents in the task recall results, comparing the fingerprint features of the documents in the task recall results with the fingerprint features of the viewed documents in the browsing history record, and filtering out documents that match the fingerprint features of the viewed documents to obtain unviewed documents.

[0139] It is worth noting that in the embodiment of the present application, the plurality of documents may only include unviewed documents, that is, when performing information search based on subscription fields, the documents obtained by the search will be filtered to include viewed documents, so that the obtained business recall results will not include viewed documents, and in this case, the unviewed documents will be directly displayed.

[0140] (2) Displaying viewed documents and unviewed documents in different areas of the same interface. For example, Fig. 7e is a schematic diagram showing how viewed documents and unviewed documents are displayed in different areas of an information search interface provided by an exemplary embodiment of the present application. In Fig. 7e, viewed documents 711 and viewed documents 712 are displayed in area 75 of the information search interface, and unviewed documents 713 and unviewed documents 714 are displayed in area 76 of the information search interface.

[0141] (3) Viewed documents and unviewed documents are displayed in different interfaces.

[0142] (4) Supports displaying unviewed documents and collapsing viewed documents in the unviewed document display interface, and supporting the collapsed viewed documents to be unfolded and displayed when triggered. In one implementation, the collapsed viewed documents correspond to a display option, and when the display option is triggered, the viewed documents can be unfolded and displayed in the display interface or a new interface independent of the display interface. For example, FIG. 7f shows a schematic diagram of viewed and unviewed documents provided by an exemplary embodiment of the present application. In FIG. 7f, the unviewed documents include document 715, which is displayed in display interface 702, and the viewed documents are collapsed in display interface 702. The collapsed viewed documents correspond to a display option 77, and when the display option 77 is triggered, the viewed documents are unfolded and displayed in a new interface 703 independent of the display interface.

[0143] (5) Displaying a plurality of documents and highlighting unviewed documents among the plurality of documents, and highlighting unviewed documents among the plurality of documents may include at least one of displaying unviewed documents in bold and displaying unviewed documents in italics, etc.

[0144] In some other possible embodiments, the multiple documents may include searched documents and recommended documents, and outputting the attention information related to the subscription field includes at least one of the following AEs: A: Hiding the searched documents and displaying only the recommended documents; B: Hiding the recommended documents and displaying only the searched documents. It should be understood that the hidden documents can be expanded and displayed when triggered. C: Categorizing and displaying the searched documents and recommended documents. For example, the searched documents can be displayed in a first display area of ​​the same interface, and the recommended documents can be displayed in a second area. D: Merging and displaying the searched documents and recommended documents. In one implementation, merging and displaying the searched documents and recommended documents may include performing a deduplication process on the searched documents and the recommended documents, obtaining the deduplication searched documents and recommended documents, and displaying the deduplication searched documents and recommended documents. For example, the searched documents may include document 1 and document 2, and the recommended documents may include document 1 and document 3. Displaying the searched documents and the recommended documents in a merged manner may involve performing a duplicate elimination process on document 1, document 2, document 1, and document 3 to obtain document 1, document 2, and document 3, and displaying document 1, document 2, and document 3.

[0145] Another implementation method is to display the retrieved document and the recommended document in a fusion manner according to a fusion display rule. The fusion display rule includes one or more of an attention fusion display rule, a subscription field relevance fusion rule, and a document diversity fusion display rule. When the fusion display rule includes an attention fusion display rule, displaying the retrieved document and the recommended document in a fusion manner according to the fusion display rule includes displaying the retrieved document and the recommended document in a fusion manner in descending order of attention. For example, if the attention of the retrieved document is greater than that of the recommended document, the retrieved document and the recommended document are displayed in a fusion manner in descending order of attention, that is, the retrieved document and the recommended document are displayed side by side in descending order of attention.

[0146] When the fusion display rule includes a subscription field relevance fusion rule, the subscription field relevance fusion rule may include a relevance with a topic corresponding to the subscription field or a relevance with a tag corresponding to the subscription field. When the subscription field relevance fusion rule includes a relevance with a topic corresponding to the subscription field, fusion-displaying the searched document and the recommended document according to the fusion display rule includes determining a relevance between the searched document and the topic corresponding to the subscription field, and determining a relevance between the recommended document and the topic corresponding to the subscription field, and fusion-displaying the searched document and the recommended document in descending order of relevance with the topic corresponding to the subscription field.

[0147] When the fusion display rule includes a document diversity fusion display rule, for example, the document diversity includes the distribution of the number of documents from the same account, and a fusion process is performed on the retrieved documents and the recommended documents according to the distribution of the number of documents from the same account, to obtain the retrieved documents and the recommended documents after the fusion process, and display the retrieved documents and the recommended documents after the fusion process. For example, assume that there are four retrieved documents, three of which belong to the same account, and the number of recommended documents is two, and these two recommended documents also belong to the same account, and the distribution of the number of documents from the same account is two or less. In this case, a fusion process can be performed on the four retrieved documents and two recommended documents according to the distribution of the number of documents from the same account, and the fusion process here can include deleting one of the retrieved documents from the three retrieved documents that belong to the same account, finally obtaining three retrieved documents and two recommended documents, and displaying the three retrieved documents and two recommended documents.

[0148] E. Screening retrieved documents and recommended documents in descending order of attention, and displaying the documents obtained by screening. One implementation is to screen retrieved documents and recommended documents in descending order of attention, obtain Y documents, and display the Y documents. For example, Y=2, and the multiple documents include retrieved document 1, retrieved document 2, and recommended document 1. The attention of retrieved document 1 is higher than that of retrieved document 2, and the attention of retrieved document 2 is higher than that of recommended document 1. Screening retrieved documents and recommended documents in descending order of attention, and obtaining two documents, namely, retrieved document 1 and retrieved document 2.

[0149] In some possible embodiments, when an update to a document related to a subscription field occurs, the attention information includes a content update notification, which is used to prompt the user that an update to a document related to the subscription field has occurred. In this embodiment, outputting attention information related to the subscription field may include at least one of the following: (1) Outputting a content update notification at the display position of the subscription field. For example, FIG. 8a is a schematic diagram of a content update notification provided by an exemplary embodiment of the present application. The content update notification includes the number of updated documents related to the subscription field. In FIG. 8a, the number of updated documents related to the subscription field "Piano Music" is displayed at the display position 81 of the subscription field "Piano Music," and the number of updated documents related to the subscription field "Piano Music" is displayed at the display position 82 of the subscription field "Cooking." (2) Outputting a content update notification around the display position of the subscription field. The periphery of the display position can be understood to be any position that is within a predetermined distance (e.g., 1 millimeter, 0.1 millimeter, etc.) from the display position of the subscription field. For example, FIG. 8b is a schematic diagram of a content update notification provided by another exemplary embodiment of the present application. The content update notification includes a recommendation reason, and in FIG. 8b, the recommendation reason "there is an update to a document related to model technology" is displayed around the display position 83 of the subscription field. (3) A third interface independent of the information search interface is used to output the content update notification. Here, the third interface may be any interface other than the information search interface on the terminal. For example, the third interface may be the main interface, standby interface, message prompt interface, information interface where the client terminal is located on the terminal, service interface, etc. of the terminal, and the embodiment of the present application is not limited thereto.

[0150] Here, the output format of the content update notification may include at least one of the following: (1) The content update notification is output in text format. For example, in both Figures 8a and 8b, the content update notification is output in text format. (2) The content update notification is output in a highlighting format. The highlighting format here may be, for example, a colored graphic format (e.g., a colored dot format), a bold format, a tilted format, or an enlarged format. For example, in Figure 8a, the content update notification is output in a bold format. (3) The content update notification is output in an audio format. When a content update notification exists, the content update notification is directly played as audio.

[0151] As described above, by actively outputting content update notifications, the subject does not need to actively check whether there are content updates in the subscription fields to which the subject has subscribed in the information search interface, which simplifies information search operations, improves information search efficiency, and better enhances the subject's experience.

[0152] Next, a related discussion will be made regarding the information processing device provided by the embodiment of the present application.

[0153] 9, which is a structural schematic diagram of an information processing device provided in an embodiment of the present application, which may be a computer program (including program code) in a computer device, for example, the information processing device may be an application software in a computer device, and the information processing device may be used to perform some or all of the steps in the embodiment of the method shown in FIG. 9. As shown in FIG. 9, the information processing device includes a display unit 901 and a processing unit 902.

[0154] The display unit 901 is used to display an information search interface, and the information search interface is used to perform information search according to a search field; The display unit 901 is further used to display a subscription field, which refers to a configured search field; The processing unit 902 is used for outputting attention information related to the subscription field.

[0155] Here, the processing unit 902 further Obtaining a subscription field, where the subscription field supports editing, where editing includes modifying or deleting; When a subscription field is changed by editing, the changed subscription field is refreshed and displayed.

[0156] Here, in the information search interface, a search subscription option is provided, and the processing unit 902 specifically: displaying a search subscription input window in response to a search subscription option being triggered; receiving a first search field entered in a search subscription entry window; and configuring the first search field as a subscription field.

[0157] Here, the information search interface includes a search field input control, and the processing unit 902 specifically: receiving a second search field entered in a search field input control; and in response to a configuration operation on the second search field, configuring the second search field as a subscription field.

[0158] Here, the processing unit 902 specifically: Displaying at least one history search field, the history search field referring to a search field used in a history information search process; and receiving a selection operation on the history search field; and configuring the selected history search field as a subscription field.

[0159] Here, in the information search interface, a search subscription option is provided, and the processing unit 902 specifically: displaying one or more suggestion search fields in response to the search subscription option being triggered; receiving a selection operation on a candidate search field; and configuring the selected candidate search field as a subscription field.

[0160] where the number of subscription fields is N, and the display unit 901 specifically includes: displaying a subscription field in an information retrieval interface; displaying a subscription field in a first interface independent of the information retrieval interface; Tiling N subscription fields; displaying the N subscription fields in order of earliest to latest configuration time; Displaying N subscription fields in order of their popularity; randomly displaying N subscription fields; Classifying and displaying the N subscription fields according to the topics to which the N subscription fields belong; Collapse and display N subscription fields; It is used to display the N subscription fields in descending order of their field lengths.

[0161] Here, the processing unit 902 specifically: outputting information of interest in a fixed area in the information search interface; outputting the information of interest in a floating area in the information search interface; and outputting the information of interest in a second interface independent of the information search interface.

[0162] Here, the attention information includes task recall results obtained by performing information search based on the subscription field, and the task recall results include a plurality of documents that match the subscription field, and the documents include sentences, web pages, videos, images, or audio. The processing unit 902 specifically: Displaying a plurality of documents in a random order; displaying the documents in order of their matching with the subscription field; Displaying multiple documents in order of their popularity; Displaying multiple documents side by side according to document diversity needs; It is used to cluster and display multiple documents according to the topic or tag to which the documents belong.

[0163] Here, the attention information includes a task recall result obtained by performing an information search based on the subscription field, the task recall result includes a plurality of documents that match the subscription field, and the plurality of documents includes viewed documents and unviewed documents. Specifically, the processing unit 902: Filtering viewed documents to display only unviewed documents; Displaying viewed and unviewed documents in different areas of the same interface; Displaying viewed and unviewed documents in different interfaces; Displaying unviewed documents and collapsing viewed documents in the display interface of the unviewed documents, wherein the collapsed viewed documents are supported to be expanded and displayed when triggered; The present invention is used to display a plurality of documents and highlight unviewed documents among the plurality of documents.

[0164] Here, the attention information includes a task recall result obtained by performing information retrieval based on the subscription field, the task recall result includes a plurality of documents that match the subscription field, and the plurality of documents includes search documents and recommended documents. The processing unit 902 specifically: Hide search documents and display only recommended documents; Hide recommended documents and display only searched documents; classifying and displaying the retrieved documents and the recommended documents; Displaying the searched document and the recommended document in a merged form; The method is used to screen retrieved documents and recommended documents in order of popularity, and to display the documents obtained by screening.

[0165] Here, the attention information includes a content update notification of the subscription field, and the processing unit 902 specifically: Outputting a content update notification at the display position of the subscription field; Outputting a content update notification around the display position of the subscription field; A third interface independent of the information search interface is adopted to output content update notifications. Here, the content update notification includes one or two of the number of updated documents and the reason for recommendation, and the output format of the content update notification includes at least one of outputting the content update notification in text format, outputting the content update notification in highlight format, and outputting the content update notification in audio format.

[0166] Here, the attention information includes a task recall result obtained by performing an information search based on the subscription field, and the task recall result includes a search recall result. The processing unit 902 further comprises: retrieving M candidate search documents matching the subscription field from the search document library within a first preset time period through the search recall link, where M is a positive integer; filtering M candidate search documents based on the target historical browsing records to obtain L candidate search documents that have not been viewed, where L is less than or equal to M; performing a sorting process on the L number of unviewed candidate search documents to obtain the L number of sorted candidate search documents; Based on the target individual data, recall processing of the individual retrieved document is performed to obtain the individual retrieved document; performing a fusion sorting process on the sorted L candidate search documents and the individual search documents to obtain a search recall result; Based on the search recall results, the recall results obtained by performing an information search based on the subscription field are obtained.

[0167] Wherein, the task recall result further includes a recommendation recall result, and the processing unit 902 further comprises: According to the subscription field, determine the topic or tag mapped to the subscription field from the recommendation recall document library provided by the recommendation recall link, where the recommendation recall information library includes at least one of candidate recommendation documents corresponding to different topics and candidate recommendation documents corresponding to different tags, and the candidate recommendation documents in the recommendation recall document library are constructed based on the top Y candidate recommendation documents located in the sorted candidate recommendation documents after sorting the candidate recommendation documents within the target time period according to the descending order of the candidate recommendation document weight; Searching a recommendation recall document library for candidate recommendation documents that match the subscription fields based on the mapped topics or tags; performing a sorting process on the matching candidate recommendation documents obtained by the search according to the sorting parameters, and obtaining the sorted candidate recommendation documents; filtering the sorted candidate recommendation documents based on the browsing history record to obtain unviewed candidate recommendation documents; According to the target information and the individual data, re-sorting the unviewed candidate recommendation documents to obtain a recommendation recall result; The search recall result and the recommendation recall result are fused to obtain a business recall result.

[0168] Wherein, the attention information includes a content update notification of a subscription field, and the processing unit 902 further: Obtaining a subscription field from the dispatch queue, and retrieving one or more local search documents that match the subscription field from a local search document library through a search recall link, wherein the publication times of the local search documents in the local search document library are all within a second preset time period; determining a topic or tag mapped to the subscription field from a recommendation recall document library provided by the recommendation recall link, and searching the recommendation recall document library for one or more candidate recommendation documents that match the subscription field based on the mapped topic or tag; performing a filtering process on one or more local search documents retrieved in the search recall link based on the history browsing record to obtain a first filtering result; performing a filtering process on one or more candidate recommendation documents retrieved by the recommendation recall link based on the browsing history record to obtain a second filtering result; performing a fusion process on the first filtering result and the second filtering result to obtain an updated recall result; If the update recall result is not an empty set, the content of the subscription field is used to generate an update notification based on the update recall result.

[0169] In an embodiment of the present application, an information search interface is displayed, which is used to perform information search based on search fields, and displays subscription fields, where the subscription fields refer to configured search fields, and outputs attention information related to the subscription fields. In this manner, not only can information search be performed based on search fields in the information search interface, but also the subject can configure the search fields as subscription fields according to their continuous attention needs, and by displaying the subscription fields, the configured search fields that need to be continuously paid attention can be presented, and the subject can know the search fields that are continuously interested, and by continuously performing information search through the subscription fields, the information that the subject is continuously interested in can be obtained and output, which enriches the information search method and at the same time improves the flexibility of information search, thereby better adapting to the scenarios in which the subject continuously pays attention to information.

[0170] Next, a related discussion will be given for the computer device provided by the embodiment of the present application.

[0171] Furthermore, the embodiment of the present application further provides a structural diagram of a computer device, which can be seen in FIG. 10 . The computer device may be the server described above, and may include a processor 1001, an input device 1002, an output device 1003, and a memory 1004. The processor 1001, the input device 1002, the output device 1003, and the memory 1004 are connected by a bus. The memory 1004 is used to store a computer program, which includes program instructions, and the processor 1001 is used to execute the program instructions stored in the memory 1004. Here, the processor 1001 can execute the information processing method described in the embodiment associated with FIG. 3A above and the information processing device described in the embodiment associated with FIG. 9 above by operating the program instructions in the memory 1004, and detailed descriptions thereof will be omitted here. Furthermore, detailed descriptions of the beneficial effects of employing the same method will also be omitted.

[0172] It should be noted here that the embodiments of the present application also provide a computer-readable storage medium, which stores a computer program, the computer program including program instructions, and a processor can execute the method in the embodiment corresponding to FIG. 3a when executing the program instructions, so detailed description is omitted here. For technical details not disclosed in the embodiments of the computer-readable storage medium related to the present application, please refer to the description of the method embodiments of the present application. For example, the program instructions can be deployed on one computer device, or executed on multiple computer devices located in one location, or executed on multiple computer devices distributed in multiple locations and interconnected by a communication network.

[0173] According to one aspect of the present application, there is provided a computer program product, the computer program product including a computer program stored in a computer-readable storage medium, a processor of a computer device reading the computer program from the computer-readable storage medium, and the processor executing the computer program to cause the computer device to perform the method in the embodiment corresponding to Fig. 3a above, and therefore a detailed description thereof will be omitted here.

[0174] The above disclosure is merely a preferred embodiment of the present application, which of course cannot limit the scope of the present application, and therefore, equivalent changes made in accordance with the claims of the present application still belong to the scope covered by the present application. [Explanation of symbols]

[0175] 31 Search field input control (component) 32 Subscription Fields 36 Subscription Field Expansion Options 51 Fixed area 52 Floating Area 61 Search Subscription Options 62 Search field input control 63 Subscription Setting Options 71 Delete Options 72 Delete Options 77 Display Options 101 terminals 102 terminals 103 Server 301 Information Search Interface 302 Information Search Interface Documents 321-324 501 Information Search Interface 601 Search Subscription Entry Window 702 Display Interface 703 Interface 711 Documents 712 documents 713 Unviewed Documents 714 Unviewed Documents 715 documents 901 Display Unit 902 Processing Unit 1001 processor 1002 Input Devices 1003 Output Devices 1004 memory

Claims

1. An information processing method, comprising: displaying an information search interface, the information search interface being used to perform an information search based on a search field; displaying a subscription field, the subscription field referring to a configured search field; and outputting information of interest related to the subscription field.

2. The method comprises: obtaining a subscription field, the subscription field supporting editing, the editing including modifying or deleting; 2. The method of claim 1, further comprising the step of: when the subscription field is changed by editing, refreshing and displaying the changed subscription field.

3. In the information search interface, a search subscription option is provided, and the step of obtaining a subscription field includes: displaying a search subscription entry window in response to the search subscription option being triggered; receiving a first search field entered in the search subscription entry window; and configuring the first search field as a subscription field.

4. In the information search interface, a search field input control is included, and the step of acquiring a subscription field includes: receiving a second search field entered in the search field input control; The method of any one of claims 1 to 3, further comprising: configuring the second search field as a subscription field in response to a configuration operation on the second search field.

5. The step of obtaining a subscription field includes: displaying at least one history search field, the history search field indicating a search field used in a history information search process; The method according to any one of claims 1 to 4, further comprising the step of: receiving a selection operation on the history search field; and configuring the selected history search field as a subscription field.

6. In the information search interface, a search subscription option is provided, and the step of obtaining a subscription field includes: displaying one or more candidate search fields in response to the search subscription option being triggered; The method according to any one of claims 1 to 5, further comprising the step of: receiving a selection operation on a candidate search field; and configuring the selected candidate search field as a subscription field.

7. The number of the subscription fields is N, and the step of displaying the subscription fields in the information search interface includes: displaying a subscription field in the information search interface; displaying a subscription field in a first interface independent of the information retrieval interface; tiling N subscription fields; displaying the N subscription fields in order of earliest configuration time; displaying the N subscription fields in order of increasing attention; randomly displaying the N subscription fields; classifying and displaying the N subscription fields according to the topics to which the N subscription fields belong; displaying the N subscription fields in a collapsed state; and a step of displaying the N subscription fields in descending order of field length.

8. The step of outputting interest information related to the subscription field includes: outputting the information of interest in a fixed area in the information search interface; outputting the information of interest in a floating area in the information search interface; and outputting the information of interest in a second interface independent of the information search interface.

9. The attention information includes a task recall result obtained by performing an information search based on the subscription field, the task recall result includes a plurality of documents that match the subscription field, the documents including text, web pages, videos, images, or audio, and the step of outputting the attention information related to the subscription field includes: displaying the plurality of documents in a random order; displaying the plurality of documents in order of their degree of relevance to the subscription field; displaying the plurality of documents in order of increasing attention; displaying the plurality of documents side-by-side according to document diversity needs; A method according to any one of claims 1 to 8, characterized in that it includes at least one of the steps of: clustering and displaying the plurality of documents according to the topic or tag to which the documents belong.

10. The attention information includes a task recall result obtained by performing an information search based on the subscription field, the task recall result includes a plurality of documents that match the subscription field, and the plurality of documents includes viewed documents and unviewed documents, and the step of outputting the attention information related to the subscription field includes: filtering the viewed documents to display only the unviewed documents; displaying the viewed documents and the unviewed documents in different areas of the same interface; displaying the viewed documents and the unviewed documents in different interfaces; displaying the unviewed documents and collapsing the viewed documents in the display interface of the unviewed documents, wherein the collapsed viewed documents are supported to be unfolded and displayed when triggered; and b) displaying the plurality of documents and highlighting unviewed documents in the plurality of documents.

11. The attention information includes a task recall result obtained by performing an information search based on the subscription field, the task recall result includes a plurality of documents that match the subscription field, and the plurality of documents include search documents and recommended documents, and the step of outputting the attention information related to the subscription field includes: hiding the retrieved documents and displaying only the recommended documents; hiding the recommended documents and displaying only the retrieved documents; a step of classifying and displaying the retrieved documents and the recommended documents; a step of displaying the retrieved document and the recommended document in a fusion manner; A method according to any one of claims 1 to 10, characterized in that it includes at least one of the steps of screening the retrieved documents and the recommended documents in order of popularity, and displaying the documents obtained by screening.

12. The attention information includes a content update notification of the subscription field, and the step of outputting the attention information related to the subscription field includes: outputting the content update notification at a display position of the subscription field; outputting the content update notification around a display position of the subscription field; and a step of outputting the content update notification by using a third interface independent of the information search interface; A method according to any one of claims 1 to 11, characterized in that the content update notification includes one or two of the number of updated documents and the reason for recommendation, and the output format of the content update notification includes at least one of outputting the content update notification in text format, outputting the content update notification in highlight format, and outputting the content update notification in audio format.

13. The attention information includes a task recall result obtained by performing an information search based on the subscription field, and the task recall result includes a search recall result, and the method includes: retrieving M candidate search documents matching the subscription field from a search document library within a first preset time period via a search recall link, where M is a positive integer; filtering the M candidate search documents based on the target historical browsing records to obtain L unviewed candidate search documents, where L is less than or equal to M; performing a sorting process on the L number of unviewed candidate search documents to obtain L number of sorted candidate search documents; A step of performing a recall process of the individual retrieved document based on the target individual data to obtain the individual retrieved document; The method of any one of claims 1 to 12, further comprising: performing a fusion sorting process on the sorted L candidate search documents and the individual search documents to obtain the search recall result.

14. The task recall result further includes a recommendation recall result, and the method further comprises: According to the subscription field, determining a topic or tag mapped to the subscription field from a recommendation recall document library provided by a recommendation recall link, wherein the recommendation recall information library includes at least one of candidate recommendation documents corresponding to different topics and candidate recommendation documents corresponding to different tags, and the candidate recommendation documents in the recommendation recall document library are constructed based on the top Y candidate recommendation documents located in the candidate recommendation documents after sorting the candidate recommendation documents within the target time period according to the descending order of the weight of the candidate recommendation documents; searching the recommendation recall document library for candidate recommendation documents that match the subscription fields based on the mapped topics or tags; a step of sorting the matching candidate recommendation documents obtained by the search according to the sorting parameters, and obtaining the sorted candidate recommendation documents; A step of filtering the sorted candidate recommendation documents based on the browsing history record to obtain unviewed candidate recommendation documents; performing a re-sorting process on the unviewed candidate recommendation documents according to the target information and individual data to obtain the recommendation recall result; The method according to any one of claims 1 to 13, further comprising: performing a fusion process on the search recall result and the recommendation recall result to obtain the task recall result.

15. The attention information includes a content update notification of the subscription field, and the method includes: obtaining the subscription field from a dispatch queue, and retrieving one or more local search documents that match the subscription field from a local search document library through a search recall link, wherein document publication times of the local search documents in the local search document library are all within a second preset time period; determining a topic or tag mapped to the subscription field from a recommendation recall document library provided by a recommendation recall link, and searching the recommendation recall document library for one or more candidate recommendation documents that match the subscription field based on the mapped topic or tag; filtering one or more local search documents retrieved in the search recall link based on the target history browsing record to obtain a first filtering result; filtering one or more candidate recommendation documents retrieved by the recommendation recall link based on the browsing history record to obtain a second filtering result; performing a fusion process on the first filtered result and the second filtered result to obtain an updated recall result; The method of any one of claims 1 to 14, further comprising: if the update recall result is not an empty set, generating a content update notification of the subscription field based on the update recall result.

16. An information processing device, comprising: a display unit; and a processing unit; The display unit is used to display an information search interface, and the information search interface is used to perform information search based on a search field; The display unit is further used to display a subscription field, and the subscription field refers to a configured search field; The information processing device, characterized in that the processing unit is used to output attention information related to the subscription field.

17. A computer device comprising: a processor adapted to execute computer programs; A computer readable storage medium on which a computer program is stored, the computer program performing the information processing method according to any one of claims 1 to 15 when executed by the processor.

18. A computer-readable storage medium having a computer program stored therein, the computer program performing the information processing method according to any one of claims 1 to 15 when executed by a processor.

19. A computer program product, comprising a computer program that, when executed by a processor, implements the information processing method of any one of claims 1 to 15.

Images