Content recommendation method and electronic equipment

Abstract

The embodiment of the invention relates to the technical field of intelligent interaction, in particular to a content recommendation method and electronic equipment, which can recommend contents better matched with a current input scene, improve the input efficiency and improve the user experience. According to the method, first content input by a user in a current input scene can be displayed, and second content suitable for the current input scene and associated with the first content is recommended at least according to the first content.

Description

Technical Field

[0001] The present invention relates to the field of intelligent interaction technology, and in particular to a content recommendation method and electronic device. Background Art

[0002] Input methods can recommend related content to users as they type. Currently, these recommendations are mostly based on input history or network semantics, which cannot adapt to the different input needs in different scenarios. Summary of the Invention

[0003] The embodiments of the present application provide a content recommendation method and electronic device, which determine a main device among multiple wearable electronic devices. Sub-devices with limited hardware capabilities can implement functions such as data sharing or message sending and receiving through the main device, further improving the user experience.

[0004] In a first aspect, an embodiment of the present application provides a content recommendation method that can display a first content input by a user in a current input scenario; then, based at least on the first content, recommend a second content that is applicable to the current input scenario and associated with the first content.

[0005] The first content is the content currently being input by the user. The second content is other content related to the first content that the input method app recommends to the user on the user interface. The first content or the second content can be in a format such as text or voice.

[0006] The method provided in the embodiment of the present application can recommend content that matches the input scenario based on the current input scenario. Different input scenarios can recommend different content, thereby recommending content that is closer to user expectations and improving user experience.

[0007] In some embodiments, before recommending the second content that is applicable to the current input scenario and associated with the first content, the type of the current input scenario may also be determined.

[0008] The type of input scenario can be a document editing scenario, a social communication scenario, an office scenario, a non-office scenario, etc. For example, a social communication scenario can specifically be chatting and communicating through instant messaging software.

[0009] In some embodiments, determining the type of the current input scene may include obtaining application type information of the current focus window and determining the type of the current input scene according to the application type information.

[0010] The focus window can be the window that has keyboard input permission at the current moment.

[0011] In some embodiments, the type of the current input scenario is determined based on the application type information. The input scenario can be determined to be a social communication scenario based on the application type information being a social communication category; and / or the input scenario can be determined to be a document editing scenario based on the application type information being an office application.

[0012] Social communication applications can include instant messaging software, as well as email, forum, Weibo and other communication software; office applications can include various text, spreadsheet or graphic processing applications, such as Word, PPT, Excel, etc.

[0013] In some embodiments, an input style suitable for the current input scenario may be determined; then, based on the first content and the input style, second content that matches the input style and is associated with the first content may be recommended.

[0014] Input styles can be humorous, relaxed, teasing, rigorous, respectful, serious, etc. Input styles can be divided according to input scenarios, that is, different input scenarios correspond to multiple styles, or they can be divided without being divided according to input scenarios.

[0015] In some embodiments, determining the input style applicable to the current input scenario can be based on the first content, identifying the input style applicable to the current input scenario; and / or displaying multiple style tags and determining the style selected by the user as the input style applicable to the current input scenario.

[0016] Based on the first content, an input style suitable for the current input scenario is identified. This can be by calling a search engine or a semantic model to identify the style of the first content currently input by the user, and then recommending content matching the style to the user.

[0017] In some embodiments, based on the first content and the input style, recommending the second content that conforms to the input style and is associated with the first content can be done by searching the local application data for target data that matches the input style based on the input style; then, obtaining the second content associated with the first content from the target data and recommending it.

[0018] For example, based on the input style, searching local application data for target data that matches the input style can be achieved by invoking a search engine, such as a local search engine. Obtaining second content associated with the first content from the target data can be achieved by invoking a pre-trained semantic model. The semantic model and search engine can be integrated into a single entity, collectively referred to as a search engine or semantic model, or separated into two independent models that perform the search and association functions, respectively.

[0019] In some embodiments, when the input scenario is determined to be a social communication scenario, the category of the communication partner may be determined. Subsequently, based at least on the first content, second content applicable to the current input scenario and associated with the first content may be recommended. This may be based on the first content and the category of the communication partner, with the second content being recommended that matches the category of the communication partner and is associated with the first content.

[0020] The category of the communication object can be one or more of the following categories: elders, leaders, colleagues, friends, family members, juniors, etc.

[0021] In some embodiments, the category of the communication object is determined, and multiple category labels can be displayed, with the category selected by the user as the category of the current communication object; then, based on the category of the communication object, target data matching the category of the communication object is searched in the local application data; from the target data, second content associated with the first content is obtained and recommended.

[0022] In a second aspect, an embodiment of the present application further provides an electronic device, comprising: a processor, wherein the processor is configured to execute a computer program or instruction in a memory to implement any of the methods described above.

[0023] In a third aspect, an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored program, wherein when the program is executed by a processor, the method as described in any one of the above items is implemented.

[0024] In a fourth aspect, an embodiment of the present application also provides a chip system, comprising: a communication interface for inputting and / or outputting data; a processor for executing a computer executable program so that a device equipped with the chip system executes any of the methods described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is an example diagram of an interface in an instant messaging scenario in the content recommendation method provided in an embodiment of the present application;

[0026] Figure 2 This is another example interface diagram for an instant messaging scenario in the content recommendation method provided in an embodiment of the present application;

[0027] Figure 3 This is another example interface diagram for an instant messaging scenario in the content recommendation method provided in an embodiment of the present application;

[0028] Figure 4 This is an example diagram of an interface in a document editing scenario in the content recommendation method provided in an embodiment of the present application;

[0029] Figure 5A flowchart of an embodiment of a content recommendation method provided in an embodiment of the present application;

[0030] Figure 6 An example diagram of an interface showing a communication object category label in the content recommendation method provided in an embodiment of the present application;

[0031] Figure 7 A schematic diagram of the architecture for searching and associating based on communication object categories in the content recommendation method provided in an embodiment of the present application;

[0032] Figure 8 A flowchart of another embodiment of the content recommendation method provided in the embodiment of the present application;

[0033] Figure 9 A schematic diagram of the architecture for performing search association based on input style in the content recommendation method provided in an embodiment of the present application;

[0034] Figure 10 An example diagram of an interface showing inputting a style tag in the content recommendation method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0035] In order to better understand the technical solutions of this specification, the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0036] It should be clear that the embodiments described are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this specification.

[0037] The terms used in the examples of this application are for the purpose of describing specific embodiments only and are not intended to limit this specification. The singular forms "a," "an," "the," and "the" used in the examples of this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0038] Input method is an encoding method used to input various symbols into electronic devices such as computers or mobile phones. Currently, some input method applications will recommend words or phrases related to the content based on the pinyin, Wubi or handwritten content typed by the user. The recommendation method is generally based on network semantic understanding or historical input records to associate recommended words to optimize the language style and recommendation intention semantics.

[0039] In practice, the same user may have different input purposes in different input scenarios and may need recommendations of different styles. For example, when chatting with friends via instant messaging, a more relaxed and humorous language style may be preferred, while when communicating with leaders via social messaging apps, a more respectful or cautious language style may be desirable. Another example is when editing technical documents, such as papers, they may need recommendations for specialized terminology within that technical field. The unified association of recommended terms based on network semantic understanding or historical input records, without distinguishing input purposes, fails to consider the diverse input needs of real-world scenarios and needs improvement.

[0040] Furthermore, since some input methods do not distinguish between input occasions or input scenarios, historical input records in a certain scenario may cause privacy leaks in other scenarios. For example, after a user uses an electronic device (such as a computer or mobile phone) to communicate with a friend through instant messaging software, they may talk about non-work content, which the user may not want to disclose in a workplace. Afterwards, the user may use the same electronic device for office work, such as starting an online meeting. In this case, if the user needs to enter some content when sharing the screen, some current input methods will search for related recommended words based on local historical input records, and may recommend non-work content that the user has entered when talking with friends, resulting in user privacy leaks and reduced user experience.

[0041] In addition, some current input method applications support personalized input, but the personalized input method only stays at the skin level.

[0042] In view of this, the embodiment of the present application proposes a content recommendation method, which can be applied to applications such as input methods, such as Figure 1 As shown, taking the instant messaging scenario as an example, the first content 10 input by the user in the current input scenario can be displayed, and then, based on the first content and the current input scenario, the second content 20 that is applicable to the current input scenario and associated with the first content is recommended.

[0043] The first content and the second content can be in text format, voice format, or other data formats that support playback or display, and support multiple natural languages. In the following description, the first content or the second content is mostly in text format and Chinese is used as an example for exemplary description.

[0044] The method provided in the embodiment of the present application can combine the current input scenario to recommend related content suitable for the current input scenario, thereby solving the user's need to input content of different styles in different scenarios.

[0045] For example, Figure 2As shown, first, the input method determines that the current input scenario is an instant messaging scenario, and then determines that the category of the communication object is "friend". Therefore, the current input scenario can specifically be an instant messaging scenario for communicating with friends. Assuming that the first content example 101 currently input by the user is the Chinese text "before", combined with the current input scenario and the category of the communication object, the second content example 201 recommended to be associated with the first content example "before" is "face", "a few days", "two days", "girlfriend", "husband" and other words or phrases. Among them, words such as "ex-girlfriend" and "ex-husband" are suitable for recommendation in instant messaging scenarios for communicating with friends, while in other scenarios, such as office scenarios or when communicating with leaders or colleagues at work, the probability of being needed is low, and therefore it is not suitable for recommendation.

[0046] For example, Figure 3 As shown, taking the instant messaging scenario as an example, when the scenario is specifically an instant messaging scenario for communicating with leaders, especially when conducting work reports or technical project docking, it is not appropriate to recommend words such as "ex-girlfriend" and "ex-husband". For the same first content example 101, the second content example 202 that can be recommended is "surface", "a few days", "two days", "end", "unprecedented" and other words or phrases. For example, words such as "front end" and "front end interaction" may have appeared in the user's previous technical reports. By querying the local application data corresponding to the category of the communication object, it can be associated with vocabulary and phrases that match or are applicable to the scenario, and "husband", "girlfriend" and so on will not be recommended as the second content.

[0047] For example, in a document editing scenario, the user is editing a technical document, and the first content (example) 102 input by the user is the English text "transf". The input method will recognize that the current input scenario is a document editing scenario, and then call the local search engine to search for application data related to "tranf" in the local application data through the search engine, and filter out words or phrases or sentence beginnings whose correlation with "tranf" meets the preset threshold from the application data as the second content recommended to the user, for example, recommending Figure 4 The second content example 203 shown specifically includes the text "transformer", "2. Transformer model is a deep learning...", "attention mechanism", "transform" and other multiple words or sentences, among which "2. Transformer model is a deep learning..." may be the content that the user has collected, annotated or marked when browsing the web recently, or it may be the data content that the user has recently added to applications such as notes or memos. In this way, the user no longer needs to open the corresponding web page or open applications such as notebooks or memos, but can enter keywords or index content through the input method to directly read or recommend related content, thereby improving input efficiency.

[0048] In the method provided in the embodiment of the present application, the type of the current input scenario may include a document editing scenario, a social communication scenario, an office scenario, a non-office scenario, etc., and may also be divided into multiple categories of application scenarios in other ways. Among them, the document editing scenario includes scenarios in which text is processed or edited using various office software such as Word and PPT. The social communication scenario may be an instant messaging scenario, such as a scenario in which communication is performed through instant messaging software, or a non-instant messaging scenario, such as a scenario in which communication is performed using email or on platforms such as forums and Weibo.

[0049] Exemplarily, the current application scenario can be determined by obtaining application type information of the currently focused window and determining the type of the current input scenario based on the application type information. For example, if the application type information indicates social communication, the input scenario is determined to be a social communication scenario. Alternatively, if the application type information indicates an office application, the input scenario is determined to be a document editing scenario or an office scenario.

[0050] Exemplarily, to obtain the application type information of the current focus window, the input method application may call an API (Application Programming Interface), which is a function that can obtain the application type information of the current focus window. By calling this function, the specific application type information of the current focus window can be obtained. Different types of applications can be identified based on the application type information, for example, whether it belongs to the instant messaging or document type, etc.

[0051] Furthermore, in some embodiments, when the current input scenario is classified as a document editing scenario or a social communication scenario, determining the input scenario and recommending second content suitable for the current input scenario can be understood as determining the input style for the current input scenario and recommending second content that matches the input style. For example, in scenarios such as document editing, after determining that the current input scenario is a document editing scenario, the input style suitable for the scenario is further determined, and then second content matching the input style is recommended.

[0052] In other embodiments, when it is determined that the category of the current input scene is a social communication scene, determining the input scene and recommending the second content applicable to the current input scene can be understood as determining the category information of the communication object under the current input scene and recommending the second content that matches or conforms to the category information of the object.

[0053] The following takes the instant messaging scenario as an example to illustrate how to implement the recommendation of related content in this scenario. Figure 5 and Figure 7The process of this method may include the following operations:

[0054] S501: Detecting first content input by a user.

[0055] S502: Determine that the current input scenario is an instant messaging scenario.

[0056] S503: Determine the category of the communication object.

[0057] The category of the communication object can be determined based on the user's operation. For example, during a chat, when the user is typing using the input method, the input method application can display multiple category labels, such as Figure 6 As shown, multiple category labels 301 (also called identity labels) such as "elders", "leaders", "colleagues", "friends", "girlfriends", "wife", "boyfriends", and "husbands" can be displayed, and users can select a label that suits the current chat object.

[0058] In other embodiments, a user-insensitive determination method can be adopted, that is, after detecting the first content input by the user, the first content is sent to a local search engine or to a semantic model, and the category of the communication object corresponding to the first content is intelligently identified through the local search engine or the semantic model. For example, if the first content is "wife", the identity of the communication object can be directly determined. For example, if the first content is "X worker", "XX general manager", "X director", the category of the communication object can be identified as a colleague, leader, or superior, etc.

[0059] Alternatively, if the first content is a name or nickname, the historical chat records corresponding to the name or nickname can be recorded in the local database in advance, and then in the subsequent conversation with the object, relevant content can be found in the historical chat records corresponding to the object for recommendation.

[0060] S504 : Recommending, based on the first content and the category of the communication partner, second content that matches the category of the communication partner and is associated with the first content.

[0061] The category of the communication object can be the identity information of a specific object or the category information of a class of communication objects. For example, object categories such as "father" and "mother" can be understood as the identity information of a specific object, and "elder" can be a category information.

[0062] After determining the category of the communication object, target data (or candidate data) matching the category of the communication object is searched in the local application data; second content associated with the first content is obtained from the target data and recommended.

[0063] For example, Figure 7As shown, after determining the communication object, an index is generated based on the first content and object information, and a call request is sent to the local search engine, and the call request carries the index information. The local search engine performs a local search based on the index, and searches the local data of multiple applications with access rights for candidate data (i.e., target data) that may be associated with the communication object, and then further filters out content with a high degree of correlation with the first content from the candidate data as the second content recommendation. The local search engine can have both search and association functions. If the local search engine does not support association, the semantic model can be called to obtain the second content with a high degree of correlation with the first content from the candidate data.

[0064] Specifically, semantic recommendation based on communication objects can also be understood as semantic recommendation based on contacts. When access to the identity information of the current communication object is obtained, a certain instance object can be automatically associated based on the user account, nickname, etc. After a successful match, the historical data of the instance (contact) is associated as a priority database. For example, when opening the chat interface of contact Zhang San in the instant messaging application, the input method will query the input method based on Zhang San's nickname or note name to see if there is historical communication information of the contact. If so, it will recommend expression style information that matches the style of the contact based on the user input content. It should be noted that some instant messaging applications may not allow access to the user's identity information. Therefore, the above-mentioned object category-based recognition method can be used to determine the category of the communication object.

[0065] The semantic model may be a natural language processing model, such as a bag of words model, a recurrent neural network model (RNN), a transformer, a GPT (Generative Pre-trained Transformer), a BERT (Bidirectional Encoder Representations from Transformers), or one or more models.

[0066] The following takes the document editing scenario as an example to illustrate how to implement the recommendation of related content in this scenario. Figure 8 and Figure 9 The process of this method may include the following operations:

[0067] S801: Detecting first content input by a user.

[0068] S802: Determine that the current input scenario is a document editing scenario.

[0069] S803: Determine the input style.

[0070] There are two ways to determine the input style that is suitable for the current input scenario:

[0071] An input style suitable for the current input scenario is identified based on the first content. For example, the first content may be sent to a local search engine or a semantic model, and the semantic model is used to intelligently identify the language style of the first content to obtain the input style.

[0072] Another way is to determine the input style based on user operations, for example Figure 10 As shown, multiple style tags 302 may be displayed, such as “humorous,” “serious,” “ridiculous,” “rigorous,” etc., and the style selected by the user may be determined as the input style suitable for the current input scenario.

[0073] S804: Recommend, based on the first content and the input style, second content that matches the input style and is associated with the first content.

[0074] Specifically, based on the first content and the input style, the second content that matches the input style and is associated with the first content is recommended. This can be done by searching for target data that matches the input style in the local application data based on the input style, and then obtaining and recommending the second content associated with the first content in the target data.

[0075] For example, Figure 9 As shown, after determining the input style, an index is generated based on the first content and the input style, and a call request is sent to the local search engine, carrying the index information. The local search engine performs a local search based on the index, searching the local data of multiple applications with access permissions for candidate data (i.e., target data) that may match the input style. The local search engine then further filters the candidate data to select content that is highly relevant to the first content and recommends it as the second content.

[0076] Specifically, for example, the input method interface displays a plurality of style tag selection items, and the user can select the "popular" style, and the input method outputs semantic recommendations of the popular style.

[0077] For example, in the above embodiments, the semantic model may include a collection module that can access a database. In some embodiments, the database may be a pre-configured database corresponding to an input method, in which data may be categorized and stored according to different communication object categories, different input styles, or different input scenarios. In other embodiments, a separate database may not be required, and instead access to cached data from multiple local applications may be requested.

[0078] The database data can be sourced from local user files, application data, and historical data caches, serving as a database of user cognition and intent. It is indexed using different instance-based methods. The instance-based method involves indexing based on input style, communication object, or scenario category.

[0079] Specifically, the data objects that the local search engine can access can come from the following sources:

[0080] Local files: When users browse (edit) information, files, etc., they may collect, mark, and tag the parts that interest them, and label the relevant information into different types. For example, different tags can be divided according to the source of the information (people, groups), or different style categories can be divided according to the style, such as language style (praise, sarcasm, youth), personal style (passion, aloofness), and literacy (elegant, professional, popular). The input method can also use semantic search to associate other content of the terminal device in real time.

[0081] Application data: The cached content of application data, or the historical communication content of the chat object can also be used as the data source of the semantic data area. For example, after obtaining access rights, the corresponding data of a user's historical communication (chat) content on the social software can be read; if the communication object is a frequently used contact or an important contact, the historical communication content of the contact can be stored as a separate category of data in the local database corresponding to the input. It should be noted that the local database corresponding to the input method can store specific instance data of various types (that is, data of a specific communication object or data of a certain input style), and can also store indexes of various instance data such as application data content.

[0082] Secondly, when searching and associating recommended content, the data object can be a single category of data or a combination of multiple categories of data. For example, one input scenario corresponds to one data category, one type of communication object corresponds to another data category, and one input style corresponds to yet another data category. The three categories of data can be searched individually or in combination. For example, some input scenarios may match both a certain input style and a certain type of communication object. In this case, a combined search can be performed in the databases corresponding to the input style and communication object.

[0083] Furthermore, for data objects of multiple categories, association priorities can be set. For example, the input method automatically prioritizes association with the corresponding instance database based on the current input scenario, input style, or chat object (i.e., communication object). Specifically, data corresponding to the input scenario, input style, or communication object can be pre-stored in the database, and priorities can be set. For example, the priority of the communication object can be set higher than the scenario, or the priority of the input style can be set higher than the priority of the communication object, and so on. In order of priority, the data with higher priority is selected for search and association.

[0084] For example, if a user needs to write a "popular" article, the input method will prioritize using popular instance data indexes to associate information that has been collected, annotated, and tagged, and recommend corresponding semantic predictions based on this type of information. For another example, when chatting with a contact on a social communication software, the input method will first query whether it has instance data for that contact. If so, it will use this as the priority database for semantic recommendations in the input box interface for that contact. For another example, if a certain input purpose or contact does not have a corresponding proprietary instance database, then the search will be performed using the normal semantic association method.

[0085] In summary, in the related technologies, semantic predictions are only made based on the user's own input records, without distinguishing the user's current application scenario or input purpose. Sometimes the recommended content is difficult to meet the user's expectations, resulting in inadequate expression of the meaning, or causing privacy leakage due to switching between different scenarios without switching the input method history records.

[0086] The method provided in the embodiment of the present application first detects the current input scene based on the first content currently input by the user, and recommends the second content that matches the current input scene and is associated with the first content.

[0087] Among them, matching with the current input scene can be understood as the scenes are divided into different categories, and matching with the current scene category; or, the current input scene is a scene that requires input according to a specified input style, and matching with the current input scene can be matching with the input style in the current scene; or, it can also be matching with the style required by the communication object in the current input scene.

[0088] It should be noted that the embodiments of the present application can implement data search and association by calling a search engine. The search engine can be a local search engine or a search engine with a network search function.

[0089] The data objects to be searched can be network data or local application data. For example, local application data can include information that users follow and collect daily, or historical chats or communication information of important contacts. Content that users have viewed, marked, or annotated is considered knowledge reserve and can be used as data objects for search.

[0090] In addition, a database dedicated to the input method can be set up to store and manage data separately in different data instance modes, that is, different instance databases can be set up to store (index) data separately and classified. In addition, users can be supported to manually or automatically switch the target database to provide the most appropriate semantic recommendation.

[0091] In this way, based on the user's different usage scenarios or expression purposes, the most appropriate data instance can be selected for matching, and automatic association based on scenario or identity matching can be achieved to make personalized (targeted) recommendations based on semantic expression. The input method can directly associate the annotated local data content based on keywords (which may be specific semantic words or a certain style).

[0092] In summary, the method provided in the embodiment of the present application can associate device-related semantic content based on local data or application history data or network data, according to user wishes and input purposes, and recommend related content, thereby bringing users a better input experience.

[0093] This method can be applied to input method applications. The resulting input method application product can select recommended content based on different users or other input purposes, which is equivalent to setting different "filters" in different scenarios and recommending content that meets the current filter, so that users do not have to worry about inappropriate expressions.

[0094] The more the recommended content fits the user's expectations, the less likely the user will need to select appropriate content from a large amount of content one by one. This can eliminate information redundancy and improve input efficiency.

[0095] An embodiment of the present application further provides an electronic device, comprising: a processor, wherein the processor is configured to execute a computer program or instruction in a memory to implement the method described in any of the above embodiments.

[0096] Exemplarily, the processor may include one or more processing units, for example, a neural-network processing unit (NPU), an application processor (AP), a modem processor, a graphics processing unit (GPU), an image signal processor (ISP), a controller, a digital signal processor (DSP), a baseband processor, etc. The different processing units may be independent devices or integrated into one or more processors. The controller may generate operation control signals based on instruction opcodes and timing signals to control instruction fetching and execution.

[0097] The memory can be used to store computer executable program code, which includes instructions. The internal memory may include a program storage area and a data storage area. Among them, the program storage area can store an operating system, an application required for at least one function, etc. The data storage area can store data (such as input data, output data) created during the use of the electronic device. In addition, the internal memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, a universal flash storage (UFS), etc. The processor executes various functional applications and data processing of the electronic device by running instructions stored in the internal memory and / or instructions stored in a memory provided in the processor.

[0098] It should be understood that the structure illustrated in the embodiment of the present invention is merely an example and does not limit the electronic device. The electronic device in the embodiment of the present application may include more or fewer components than shown, or combine or separate certain components, or arrange the components differently. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0099] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored program, wherein when the program is executed by a processor, the method described in any of the above embodiments is implemented.

[0100] An embodiment of the present application further provides a computer program product, which includes a program. When the program is executed by an electronic device, the electronic device implements the method described in any of the above embodiments.

[0101] An embodiment of the present application also provides a chip system, including: a communication interface for inputting and / or outputting data; and a processor for executing a computer executable program so that a device equipped with the chip system executes a method as described in any of the above embodiments.

[0102] The above-mentioned computer-readable storage medium can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination thereof. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM) or flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or device.

[0103] A computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0104] Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0105] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0106] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of such features. Throughout the description of this application, "plurality" means at least two, for example, two, three, etc., unless otherwise specifically defined.

[0107] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present application belong.

Claims

1. A content recommendation method, characterized in that: The method comprises: Display the first content entered by the user in the current input scenario; Based at least on the first content, second content that is applicable to the current input scenario and associated with the first content is recommended.

2. The method according to claim 1, wherein Before recommending second content that is applicable to the current input scenario and associated with the first content, the method further includes: Determines the type of the current input scene.

3. The method according to claim 2, wherein Determine the type of the current input scene, including: Obtain application type information of the current focus window, and determine the type of the current input scenario based on the application type information.

4. The method according to claim 3, wherein Determining the type of the current input scenario based on the application type information includes: According to the application type information being a social communication type, determining that the input scenario is a social communication scenario; and / or, According to the application type information being an office application, it is determined that the input scenario is a document editing scenario.

5. The method according to any one of claims 1 to 4, wherein The method further comprises: Determine the input style appropriate for the current input scenario; Recommending, based at least on the first content, second content that is applicable to the current input scenario and associated with the first content, includes: According to the first content and the input style, second content that matches the input style and is associated with the first content is recommended.

6. The method according to claim 5, wherein Determine the input style appropriate for the current input scenario, including: Based on the first content, identifying an input style suitable for the current input scenario; and / or, Multiple style tags are displayed, and the style selected by the user is determined as the input style applicable to the current input scenario.

7. The method according to claim 5 or 6, wherein: Recommending, based on the first content and the input style, second content that matches the input style and is associated with the first content, includes: Searching, according to the input style, the local application data for target data that matches the input style; Second content associated with the first content is obtained from the target data and recommended.

8. The method according to claim 4, wherein When the input scenario is determined to be a social communication scenario, before recommending, based at least on the first content, second content applicable to the current input scenario and associated with the first content, the method further includes: Determine the category of recipients of communications; Recommending, based at least on the first content, second content that is applicable to the current input scenario and associated with the first content, includes: According to the first content and the category of the communication object, second content that matches the category of the communication object and is associated with the first content is recommended.

9. The method according to claim 8, wherein Identify the categories of communication recipients, including: Display multiple category labels and use the category selected by the user as the category of the current communication object; Recommending, based on the first content and the category of the communication object, second content that matches the category of the communication object and is associated with the first content, includes: According to the category of the communication object, searching the local application data for target data that matches the category of the communication object; Second content associated with the first content is obtained from the target data and recommended.

10. An electronic device, characterized in that: The electronic device comprises: A processor, configured to execute a computer program or instruction in a memory to implement the method according to any one of claims 1 to 9.

11. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored program, wherein the program implements the method according to any one of claims 1 to 9 when executed by a processor.

12. A chip system, characterized in that: include: a communication interface for inputting and / or outputting data; A processor, configured to execute a computer executable program so that a device equipped with the chip system executes the method according to any one of claims 1 to 9.

Images