Recommended language updating method and apparatus, device, storage medium and product

Abstract

Embodiments of the present application provide a recommended language updating method and apparatus, a device, a storage medium and a product. The technical solution provided by an embodiment of the present application comprises: on the basis of user input data, determining one or more input languages and an occurrence count of each input language; on the basis of keyboard usage data, determining one or more keyboard languages and an occurrence count of each keyboard language; on the basis of the input languages, the occurrence count of each input language, the keyboard languages and the occurrence count of each keyboard language, determining alternate languages and alternate language scores corresponding to the alternate languages; and updating recommended languages on the basis of the alternate languages and the alternate language scores. The recommended languages are closer to a behavior habit of a user, the requirement of the user for the multi-language content is better met, and the information recommendation effect is improved.

Description

A method, device, equipment, storage medium and product for updating recommended languages

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on December 15, 2023, with application number 202311739846.2, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The embodiments of the present application relate to the field of computer technology, and in particular to a method, apparatus, device, storage medium, and product for updating recommended languages. Background Art

[0003] With the development of network and computer technology, the content consultation faced by people is becoming more and more diverse. In order to improve the efficiency of people obtaining information of interest, information content is generally recommended to users based on the content of their interest.

[0004] In international mobile applications, multilingual users often need content recommendations in different languages. However, current content recommendation methods generally base recommendations on the user's language settings, which results in relatively limited content recommendations. For example, in some multilingual regions, users may encounter content in multiple languages. Language recommendations based solely on IP addresses or user language settings cannot meet users' needs for multilingual content, resulting in poor information recommendation results. Summary of the Invention

[0005] The embodiments of the present application provide a method, apparatus, device, storage medium, and product for updating recommended languages ​​to address the problems in related technologies of a single method for determining recommended languages ​​and poor information recommendation effects. The method can flexibly update recommended languages ​​and improve information recommendation effects.

[0006] In a first aspect, an embodiment of the present application provides a method for updating a recommended language, comprising:

[0007] Determining one or more input languages ​​and the number of input times corresponding to each input language according to user input data;

[0008] Determining one or more keyboard languages ​​and the number of keyboard language usages corresponding to each of the keyboard languages ​​based on the keyboard usage data;

[0009] Determining a backup language and a backup language score corresponding to the backup language according to the input language, the number of times the language is input, the keyboard language, and the number of times the keyboard language is input;

[0010] The recommended language is updated according to the backup language and the backup language score.

[0011] In a second aspect, an embodiment of the present application provides a recommended language updating device, comprising an input language module, a keyboard language module, a backup language module, and a backup language module, wherein:

[0012] The input language module is configured to determine one or more input languages ​​and the number of input times corresponding to each input language according to user input data;

[0013] The keyboard language module is configured to determine one or more keyboard languages ​​and the number of keyboard language usages corresponding to each of the keyboard languages ​​according to the keyboard usage data;

[0014] The backup language module is configured to determine a backup language and a backup language score corresponding to the backup language based on the input language, the number of times the language is input, the keyboard language, and the number of times the keyboard language is input;

[0015] The language updating module is configured to update the recommended language according to the standby language and the standby language score.

[0016] In a third aspect, an embodiment of the present application provides a recommended language updating device, comprising: a memory and one or more processors;

[0017] The memory is used to store one or more programs;

[0018] When the one or more programs are executed by the one or more processors, the one or more processors implement the recommended language updating method as described in the first aspect.

[0019] In a fourth aspect, an embodiment of the present application provides a non-volatile storage medium storing computer-executable instructions, wherein the computer-executable instructions, when executed by a computer processor, are used to perform the recommended language updating method as described in the first aspect.

[0020] In a fifth aspect, an embodiment of the present application provides a computer program product, which includes a computer program stored in a computer-readable storage medium. At least one processor of a device reads and executes the computer program from the computer-readable storage medium, so that the device performs the recommended language update method as described in the first aspect.

[0021] The embodiments of the present application determine one or more input languages ​​and the number of input language uses corresponding to each input language based on user input data, and determine one or more keyboard languages ​​and the number of keyboard language uses corresponding to each keyboard language based on keyboard usage data, determine a backup language and a backup language score corresponding to the backup language based on the input language, the number of input language uses, the keyboard language, and the number of keyboard language uses, and update the recommended language based on the backup language and the backup language score. By determining the backup language and the corresponding backup language score based on user input data and keyboard usage data that reflect user behavior, the recommended language can be dynamically corrected to achieve flexible updating of the recommended language. The recommended language is more closely aligned with the user's behavioral habits, better meets the user's demand for multilingual content, and improves the effect of information recommendation. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] FIG1 is a flow chart of a method for updating recommended languages ​​provided in an embodiment of the present application;

[0023] FIG2 is a flow chart of another method for updating recommended languages ​​provided in an embodiment of the present application;

[0024] FIG3 is a schematic diagram of a flow chart of a backup language determination process provided by an embodiment of the present application;

[0025] FIG4 is a schematic diagram of a recommended language update process provided by an embodiment of the present application;

[0026] FIG5 is a schematic diagram of the structure of a recommended language updating device provided in an embodiment of the present application;

[0027] FIG6 is a schematic structural diagram of a recommended language updating device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and advantages of the present application clearer, the specific embodiments of the present application are further described in detail below in conjunction with the accompanying drawings. It is understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. It should also be noted that, for ease of description, only some, but not all, of the contents related to the present application are shown in the accompanying drawings. Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe each operation (or step) as a sequential process, many of the operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of the operations can be rearranged. The above process can be terminated when its operation is completed, but it can also have additional steps not included in the accompanying drawings. The above process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0029] The recommended language updating method provided in this application can be applied to information recommendation in multiple languages. For example, in a short video recommendation queue, based on the dynamically updated recommended language, the information recommendations to users are adjusted according to different user language preferences. For example, the language type of short videos and / or post content in the short video recommendation queue is adjusted. The purpose is to determine the backup language and the corresponding backup language score based on user input data and keyboard usage data that reflect user behavior, so as to dynamically correct the recommended language and realize flexible updating of the recommended language. The recommended language is closer to the user's behavioral habits, better meets the user's demand for multilingual content, and improves the information recommendation effect.

[0030] In existing information recommendation solutions, the recommended language is generally fixed to the language used by the user's device and the language used by the client. The recommended language is single and fixed, and the information content recommended to the user is all in a fixed language, resulting in poor information recommendation results. At the same time, traditional multilingual content recommendation systems usually rely solely on the user's language settings or IP address to select the appropriate language for information recommendation. However, this type of information recommendation captures the diverse needs of users, especially in situations involving cross-regional work or multilingual needs. The information recommendation effect is poor. Based on this, a method for updating recommended languages ​​in an embodiment of the present application is provided to solve the problem that the existing method for determining recommended languages ​​is single and the information recommendation effect is poor.

[0031] FIG1 is a flowchart of a method for updating recommended languages ​​provided in an embodiment of the present application. The method for updating recommended languages ​​provided in an embodiment of the present application can be executed by a recommended language updating apparatus, which can be implemented in hardware and / or software and integrated into a recommended language updating device (e.g., a server).

[0032] The following description will be made using an example of a method for updating a recommended language executed by a recommended language updating device. Referring to FIG1 , the method for updating a recommended language includes:

[0033] S110: Determine one or more input languages ​​and the number of input times corresponding to each input language according to user input data.

[0034] S120: Determine one or more keyboard languages ​​and the number of keyboard language usages corresponding to each keyboard language according to the keyboard usage data.

[0035] In one embodiment, the recommended language updating device provided by this solution can collect user behavior data from each user at set time intervals (e.g., one day, three days, one week, etc.), and execute a recommended language updating method based on the collected user behavior data to update the recommended language for each user. The user behavior data provided by this solution can be collected based on various user interactions on a platform or application (e.g., a short video platform or application) (e.g., browsing behavior, communication behavior, liking behavior, etc.) and / or interactions with user devices (e.g., mobile phones, tablets, computers, etc.) (e.g., using a system keyboard or a third-party keyboard to enter text), for example, collecting text data generated by users during these interactions.

[0036] The user behavior data provided by this solution includes user input data and keyboard usage data. The user input data can reflect the text data input by the user when performing interactive operations on the platform or application, and the keyboard usage data can reflect the user's keyboard usage. Optionally, the client installed on the user's device generates corresponding user input data or keyboard usage data each time the user performs an interactive operation on the platform or application, or an interactive operation on the keyboard, and uploads it to a set location (recommended language update device or setting database) in real time or at a time interval, or the recommended language update device generates corresponding user input data or keyboard usage data each time the user performs an interactive operation on the platform or application, or an interactive operation on the keyboard. Optionally, the input language and keyboard language provided by this solution can be represented by a language code. For example, the language corresponding to "es:30" is Spanish, and the corresponding number of times is 30.

[0037] Exemplarily, to update a user's recommended language, user input data corresponding to a set period of time (e.g., 30-60 days) is obtained, and one or more input languages ​​corresponding to the user are determined based on the user input data, as well as the number of input language times corresponding to each of the one or more input languages. The number of input language times can be understood as the number of times the corresponding input language appears. Optionally, the input language can be determined by performing language recognition on the user input data, and the number of input language times corresponding to each input language can be the number of times the user input data recognizes the corresponding input language. For example, when an input language is detected in a user's input data, the number of input language times corresponding to the input language is increased by one. Optionally, assuming that the input languages ​​detected are Spanish and Arabic, and the corresponding number of input language times are 30 and 19, respectively, the set of the user's input languages ​​and input language times can be expressed as: {es:30,ar:19}, where es and ar are the language codes corresponding to Spanish and Arabic, respectively.

[0038] Optionally, after determining one or more input languages ​​and the number of input language uses corresponding to each input language, the input languages ​​with a set number of input language uses (e.g., the top 20) are filtered out, and the input languages ​​most affected by user behavior are retained to improve the accuracy of recommended language updates.

[0039] In one embodiment, keyboard usage data corresponding to a set time length (e.g., 30-60 days) is obtained, and one or more keyboard languages ​​corresponding to the user are determined based on the keyboard usage data, and the keyboard language count corresponding to each of the one or more keyboard languages ​​is determined. The keyboard language count can be understood as the number of times the corresponding keyboard language appears. Optionally, the keyboard language can be determined based on the language corresponding to the keyboard used by the user within a set time range (e.g., 1-3 days), and the keyboard language count of each keyboard language can be the number of times or days the keyboard language appears within the set time length.

[0040] Exemplarily, the keyboard language used when using the keyboard within a set time range is collected, and the corresponding keyboard usage data within the set time range records the keyboard language. When determining one or more keyboard languages ​​and the number of keyboard language uses corresponding to each keyboard language based on the keyboard usage data, determine the various keyboard languages ​​that appear in these keyboard usage data within the set time length, and determine the number of times each keyboard language appears as the corresponding number of keyboard language uses. For example, the keyboard usage data is recorded on a daily basis, and the keyboard language used on that day is recorded in the keyboard usage data recorded each day. The corresponding keyboard usage data of the user within 30 days is obtained, the keyboard languages ​​that appear in these keyboard usage data are determined, and the number of times each keyboard language appears is used as the corresponding number of keyboard language uses.

[0041] Optionally, after determining one or more keyboard languages ​​and the number of keyboard language calls corresponding to each keyboard language based on keyboard usage data, a set number (for example, two) of keyboard languages ​​with the highest number of keyboard language calls can be retained, and multiple keyboard languages ​​that are most affected by user behavior can be retained to improve the accuracy of recommended language updates.

[0042] For example, when the keyboard language set is represented as {ar:16, bn:10}, the detected keyboard languages ​​are Arabic and Bengali, and the corresponding keyboard language times are 16 and 10 respectively. When the keyboard language set is represented as {bn:15}, the detected keyboard language is Bengali, and the corresponding keyboard language time is 15.

[0043] S130: Determine a backup language and a backup language score corresponding to the backup language according to the input language, the number of times the language is input, the keyboard language, and the number of times the keyboard is input.

[0044] Exemplarily, after determining the input language and the corresponding number of input language attempts, as well as the keyboard language and the corresponding number of keyboard language attempts, the backup language and the corresponding backup language score are determined based on the input language, the number of input language attempts, the keyboard language, and the number of keyboard language attempts.

[0045] The backup language provided by this solution can be understood as a candidate language that can be used to update the recommended language currently corresponding to the user. Optionally, based on the correspondence between the keyboard language and the input language (i.e., whether the keyboard language exists in the input language), and / or the number of input languages ​​and the number of keyboard language uses, the backup language can be determined from the keyboard language and the input language, and the corresponding language score can be determined as the backup language score. For example, among the keyboard languages ​​corresponding to the input language, the keyboard language with the largest number of keyboard languages ​​can be used as the backup language, or the keyboard language or input language with the largest number of keyboard languages ​​and the number of input languages ​​can be used as the backup language.

[0046] S140: Update the recommended language according to the backup language and the backup language score.

[0047] Exemplarily, the recommended language is updated based on the determined alternative languages, alternative language scores, recommended languages, and recommended language scores corresponding to the recommended languages. Optionally, the recommended language scores can be updated by adding new recommended languages ​​or replacing the original recommended language with the lowest score.

[0048] In one embodiment, after determining a backup language, if no corresponding recommended language currently exists, the backup language is directly determined as the recommended language. Optionally, if the number of recommended languages ​​does not reach a set number of languages ​​(e.g., two), the backup language can be directly added as a new recommended language. Alternatively, if the current recommended language score reaches a set score threshold, the recommended language only has invalid languages, and the backup language score reaches a set score threshold, the backup language can be added as a new recommended language. Alternatively, if the current recommended language score is less than a set score threshold, the backup language can be added as a new recommended language. When the number of recommended languages ​​reaches the set number of languages, a determination can be made based on the recommended language with the lowest recommended language score and the backup language score whether to replace the recommended language with the lowest recommended language score with the backup language. For example, if the recommended language score is less than a first set score threshold, or if the backup language score is greater than the minimum recommended language score and greater than a second set score threshold (greater than the first score threshold), the recommended language with the lowest recommended language score can be replaced with the backup language.

[0049] As described above, by determining one or more input languages ​​and the number of input language uses corresponding to each input language based on user input data, and determining one or more keyboard languages ​​and the number of keyboard language uses corresponding to each keyboard language based on keyboard usage data, determining backup languages ​​and backup language scores corresponding to the backup languages ​​based on the input language, the number of input language uses, the keyboard language, and the number of keyboard language uses, and updating the recommended languages ​​based on the backup languages ​​and the backup language scores, and by determining backup languages ​​and the corresponding backup language scores based on user input data and keyboard usage data that reflect user behavior, the recommended languages ​​can be dynamically corrected, and flexible updates of the recommended languages ​​can be achieved. The recommended languages ​​are more closely aligned with user behavior habits, better meet user needs for multilingual content, and improve information recommendation effects.

[0050] Based on the above embodiment, FIG2 shows a flowchart of another method for updating recommended languages ​​provided by an embodiment of the present application. This method for updating recommended languages ​​is a specific embodiment of the above method for updating recommended languages. Referring to FIG2 , this method for updating recommended languages ​​includes:

[0051] S210: Obtain user input data, and perform text preprocessing on the user input data to obtain user input text data.

[0052] Exemplarily, user input data is obtained based on pre-defined data acquisition channels. In one embodiment, the data acquisition channels provided by this solution include a description data acquisition channel, a comment data acquisition channel, a nickname data acquisition channel, a signature data acquisition channel, and a chat data acquisition channel. In one embodiment, the user input data provided by this solution includes a combination of one or more of description data, comment data, nickname data, signature data, and chat data.

[0053] The description data provided by this solution can be understood as the description of the content posted by users on platforms (such as video publishing platforms, forums, etc.) or applications (such as video applications, social applications, etc.) (the content may include text and / or non-text content, such as emoticons, punctuation marks, etc.). For example, description data is generated based on the description of the video content or related introduction content when the user posts the video content, or the content corresponding to the posted post. Comment data can be the comment text of the user on the platform or application for their own or others' posted content or live broadcast content, such as the content of comments on posts posted by others in the comment area of ​​others' videos, or the content of comments in the live broadcast room of the anchor. Nickname data is the nickname text corresponding to the account nickname used by the user. For example, when it is detected that the user has modified their account nickname, the corresponding nickname data is generated based on the modified account nickname. Signature data is the signature content filled in by the user in the signature area of ​​their user account. For example, when it is detected that the user has modified the signature content corresponding to the signature area, the corresponding signature data is generated based on the modified signature content. Chat data can be the chat content entered by the user in a personal chat or a group chat. For example, when it is detected that the user is chatting, the corresponding chat data is generated based on the chat content entered by the user. Optionally, the user input data may be collected and uploaded in real time by the user device.

[0054] In one embodiment, after obtaining user input data within a set time period, the user input data is subjected to text preprocessing to obtain user input text data. Optionally, the text preprocessing provided by this solution includes one or more combinations of punctuation removal, multi-space deduplication, lowercase normalization (converting uppercase characters to lowercase characters), emoticon removal, normalization, and full-width to half-width conversion (converting full-width to half-width).

[0055] Optionally, punctuation marks can be removed from user input data by excluding Punctuation using unicodedata, emoticons can be removed from user input data using an emoji package, and user input data can be normalized based on Unicode's NFD normalization and NFKD compatibility normalization methods. After text preprocessing of the user input data, if no characters exist in the user input data, language recognition will not be performed on the user input data, thus reducing unnecessary data processing, reducing the impact of invalid languages ​​identified based on user input data without characters on the determination of alternative languages, and improving the accuracy of recommended language updates.

[0056] In one embodiment, after obtaining user input data within a set time length, user input data with text-related content can be first screened out, and then these user input data with text-related content can be subjected to text preprocessing to obtain user input text data. Optionally, a Unicode classification method (Unicode classification method) can be used to identify whether each user input data contains text content. For example, by calling the Unicode data packet (Unicodedata package) of the user input data, the Unicode classification (Unicode classification) of the user input data is determined. If there is no text category (Letter category) in the Unicode classification of the user input data, it can be determined that the user input data has no text-related content (for example, the user input data are all punctuation marks, numbers, emoticons, etc.). If there is a text category (Letter category) in the Unicode classification of the user input data, it can be determined that the user input data has text-related content. By performing text preprocessing on the filtered user input data with text-related content to obtain user input text data, the efficiency of input language recognition is effectively improved.

[0057] S220: Perform language recognition on the text data input by the user to obtain one or more input languages, and determine the number of input languages ​​corresponding to each input language.

[0058] Exemplarily, based on a set language recognition algorithm or model, the language of the user input text determined above is recognized to obtain the input language corresponding to each user input text data, and the number of input language counts corresponding to each input language is determined based on the number of recognition times for each input language. Optionally, the language of the user input text data can be recognized based on a fast text classification model (FastText model), a TextCNN model, a TextRNN model, a TextRCNN model, a BiLSTM_Attention model, or the like.

[0059] This solution obtains user-input text data by performing text preprocessing on the user-input data, and performs language recognition on the user-input text data to obtain one or more input languages ​​and the corresponding number of times the language is input. The preprocessed user-input text data is more standardized, and the impact of non-text content on language recognition is reduced. The language recognition of the user-input text data is more accurate, which effectively improves the accuracy of recommended language updates, makes recommended content more personalized, meets the diverse needs of users, and improves user satisfaction.

[0060] S230: Determine one or more keyboard languages ​​and the number of keyboard language usages corresponding to each keyboard language according to the keyboard usage data.

[0061] S240: Determine a backup language and a backup language score corresponding to the backup language according to the number of keyboard languages ​​and the validity of the keyboard languages, as well as the input language, the number of inputs, the keyboard language, and the number of keyboard inputs.

[0062] In one embodiment, after determining the input language, the number of input language inputs, the keyboard language, and the number of keyboard language inputs, the number of keyboard languages ​​and the validity of the keyboard language are determined. Among them, the validity of the keyboard language can reflect whether the keyboard language is a valid language or an invalid language. The invalid languages ​​provided in this solution can be understood as languages ​​without clear language information and languages ​​with strong applicability (such as English, which has little effect on the user's preferred language analysis due to its strong applicability), such as English keyboard languages, emoticons, strokes, etc. This solution uses languages ​​with strong applicability as invalid languages, such as Chinese, English, etc. This solution uses English as an example of setting an invalid language. Optionally, different countries or regions (countries or regions with different languages ​​with strong applicability) can set different invalid languages.

[0063] Exemplarily, after determining the number of keyboard languages ​​and the validity of the keyboard languages, the backup languages ​​and the backup language scores corresponding to the backup languages ​​are determined from the keyboard languages ​​and input languages ​​based on the input language, the number of input languages, the keyboard language, the number of keyboard languages, and the validity of the keyboard language.

[0064] For example, if the keyboard language with the most keyboard language times is a valid language and corresponds to the input language, the keyboard language with the most keyboard language times will be selected as an alternative language, and the alternative language score will be the sum of the corresponding keyboard language times and the input language times. If the keyboard language with the most keyboard language times is a valid language but does not correspond to the input language, the input language with the most input language times will be selected as an alternative language, and the alternative language score will be the corresponding input language times. If the keyboard language with the most keyboard language times is an invalid language but the keyboard language with the second most keyboard language times is a valid language, if the keyboard language with the second most keyboard language times corresponds to the input language, the keyboard language with the second most keyboard language times will be selected as an alternative language, and the alternative language score will be the sum of the corresponding keyboard language times and the input language times. If the keyboard language with the second most keyboard language times does not correspond to the input language, the keyboard language with the second most keyboard language times will be selected as an alternative language, and the alternative language score will be the corresponding keyboard language times. This solution improves the accuracy of recommended language updates by accurately determining backup languages ​​and corresponding backup language scores based on the input language, number of inputs, keyboard language, number of keyboard language inputs, number of keyboard languages, and the validity of the keyboard language. This helps the system to more accurately correct and adjust the user's language preferences, thereby providing more appropriate multilingual content recommendations, making users more satisfied with the application's content recommendations and experience, and effectively improving user satisfaction.

[0065] In one possible embodiment, when the number of keyboard languages ​​is greater than one, the keyboard language with the most keyboard language usage is defined as the first keyboard language, and the keyboard language with the second most keyboard language usage is defined as the second keyboard language. As shown in FIG3 , a flowchart of a backup language determination process is provided. The recommended language update method provided by this solution determines the backup language and the backup language score corresponding to the backup language based on the number of keyboard languages ​​and the validity of the keyboard languages, as well as the input language, the number of input language usages, the keyboard language, and the number of keyboard language usages. The method includes:

[0066] S241: When the number of keyboard languages ​​is greater than 1 and the first keyboard language with the largest number of keyboard language uses is a valid language, determine a backup language and a backup language score corresponding to the backup language based on a correspondence between the first keyboard language and the input language, as well as the number of input language uses and the number of keyboard language uses.

[0067] For example, when the number of keyboard languages ​​is greater than 1 (for example, the number of keyboard languages ​​is 2) and the first keyboard language with the most keyboard language usage is a valid language, the correspondence between the first keyboard language and the input language is determined (i.e., whether the first keyboard language exists among the multiple input languages ​​is determined). Based on the correspondence between the first keyboard language and the input language, as well as the number of input languages ​​and the number of keyboard language usages, a backup language and its corresponding backup language score are determined. This solution effectively improves the accuracy of recommended language updates by accurately determining the backup language and its corresponding backup language score based on the correspondence between the first keyboard language and the input language, as well as the number of input languages ​​and the number of keyboard language usages, when the number of keyboard languages ​​is greater than 1 and the first keyboard language is a valid language.

[0068] In one possible embodiment, the method for updating recommended languages ​​provided by this solution, wherein determining the backup language and the backup language score corresponding to the backup language based on the correspondence between the first keyboard language and the input language, as well as the number of input language times and the number of keyboard language times, includes:

[0069] S2411: If the first keyboard language exists in the input languages, determine the first keyboard language as a standby language, and determine a standby language score corresponding to the standby language based on the sum of the keyboard language count of the first keyboard language and the input language count of the corresponding input language.

[0070] S2422: If the first keyboard language does not exist in the input languages, determine the input language with the largest number of input times as the backup language, and determine a backup language score corresponding to the backup language according to the number of input times of the input language.

[0071] Exemplarily, when the number of keyboard languages ​​is greater than 1 and the first keyboard language is a valid language, the correspondence between the first keyboard language and the input language is determined, that is, whether the first keyboard language exists in all input languages ​​is determined.

[0072] In one embodiment, when a first keyboard language exists in the input languages, the first keyboard language is determined as a standby language, and the sum of the keyboard language count of the first keyboard language and the input language count of the corresponding input language is used as the standby language score corresponding to the standby language.

[0073] In one embodiment, when the first keyboard language does not exist in the input languages, the input language with the largest number of input times is determined as the standby language, and the number of input times of the input language with the largest number of input times is determined as the standby language score corresponding to the standby language.

[0074] This solution effectively improves the accuracy of recommended language updates by accurately determining the backup language and the corresponding backup language score based on the correspondence between the first keyboard language and the input language when the number of keyboard languages ​​is greater than 1 and the first keyboard language with the most keyboard languages ​​is a valid language.

[0075] S242: When the number of keyboard languages ​​is greater than 1, the first keyboard language with the largest number of keyboard languages ​​is an invalid language, and the second keyboard language with the second largest number of keyboard languages ​​is a valid language, determine a backup language and a backup language score corresponding to the backup language based on the correspondence between the second keyboard language and the input language, as well as the number of input languages ​​and the number of keyboard languages.

[0076] For example, when the number of keyboard languages ​​is greater than one, the first keyboard language with the most keyboard language usage is an invalid language, and the second keyboard language with the second most keyboard language usage is a valid language, the backup language and its corresponding backup language score are determined based on the correspondence between the second keyboard language and the input language (i.e., determining whether the second keyboard language exists among the multiple input languages), as well as the number of input languages ​​and the number of keyboard language usages. This solution effectively improves the accuracy of recommended language updates by accurately determining the backup language and its corresponding backup language score based on the correspondence between the second keyboard language and the input language, as well as the number of input languages ​​and the number of keyboard language usages, when the number of keyboard languages ​​is greater than one, the first keyboard language is an invalid language, and the second keyboard language is a valid language.

[0077] In one possible embodiment, the recommended language updating method provided by this solution, when determining the backup language and the backup language score corresponding to the backup language based on the correspondence between the second keyboard language and the input language, as well as the number of input language times and the number of keyboard language times, includes:

[0078] S2421: If a second keyboard language exists in the input languages, determine the second keyboard language as a standby language, and determine a standby language score corresponding to the standby language based on the sum of the keyboard language count of the second keyboard language and the input language count of the corresponding input language.

[0079] S2422: If the second keyboard language does not exist in the input languages ​​and the keyboard language count of the second keyboard language is greater than a first set count, determine the second keyboard language as a backup language, and determine a backup language score corresponding to the backup language based on the keyboard language count of the second keyboard language.

[0080] S2423: If the second keyboard language does not exist in the input languages, the keyboard language count of the second keyboard language is less than or equal to the first set count, and the first keyboard language is an invalid language and the keyboard language count is greater than the second set count, the first keyboard language is determined as a backup language, and a backup language score corresponding to the backup language is determined based on the keyboard language count of the first keyboard language.

[0081] S2424: If the second keyboard language does not exist in the input languages, the keyboard language count of the second keyboard language is less than or equal to the first set count, and the first keyboard language is a non-set invalid language and / or the keyboard language count of the first keyboard language is less than or equal to the second set count, the second keyboard language is determined as a backup language, and a backup language score corresponding to the backup language is determined based on the keyboard language count of the second keyboard language.

[0082] Exemplarily, when the number of keyboard languages ​​is greater than 1, the first keyboard language with the largest number of keyboard languages ​​is an invalid language, and the second keyboard language with the second largest number of keyboard languages ​​is a valid language, it is determined whether the second keyboard language corresponds to the input language, that is, whether the second keyboard language exists in the input language.

[0083] In one embodiment, when a second keyboard language exists in the input languages, the second keyboard language is determined as a standby language, and the sum of the keyboard language count of the second keyboard language and the input language count of the corresponding input language is used as the standby language score corresponding to the standby language.

[0084] If the second keyboard language does not exist in the input language, it is determined whether the keyboard language count of the second keyboard language is greater than a first set count (e.g., the first set count is 3). In one embodiment, if the keyboard language count of the second keyboard language is greater than the first set count, the second keyboard language is determined as a backup language, and the keyboard language count of the second keyboard language is used as the backup language score corresponding to the backup language.

[0085] In one embodiment, when the keyboard language number of the second keyboard language is less than or equal to the first set number, it is determined whether the first keyboard language is an invalid language and whether the keyboard language number of the first keyboard language is greater than the second set number (the second set number is greater than the first set number, for example, the second set number is 10).

[0086] In one embodiment, when the first keyboard language is an invalid language (e.g., English) and the number of keyboard language uses of the first keyboard language is greater than a second set number, the first keyboard language is determined as a standby language, and the number of keyboard language uses of the first keyboard language is used as a standby language score corresponding to the standby language.

[0087] In one embodiment, when the first keyboard language is not a set invalid language (i.e., an invalid language without clear language information other than the set invalid language) and / or the keyboard language count of the first keyboard language is less than or equal to a second set count, the second keyboard language is determined as a backup language, and a backup language score corresponding to the backup language is determined based on the keyboard language count of the second keyboard language.

[0088] This solution determines a backup language and a backup language score corresponding to the backup language based on the correspondence between the second keyboard language and the input language, as well as the number of input languages ​​and the number of keyboard languages ​​when the number of keyboard languages ​​is greater than one, the first keyboard language is an invalid language, and the second keyboard language is a valid language.

[0089] S243: When the number of keyboard languages ​​is 1 and the keyboard language is a valid language, determine a backup language and a backup language score corresponding to the backup language according to the correspondence between the keyboard language and the input language, the number of input languages ​​and the number of keyboard languages.

[0090] For example, when the number of keyboard languages ​​is one and the keyboard language is a valid language, the backup language and its corresponding backup language score are determined based on the correspondence between the keyboard language and the input language (i.e., determining whether the keyboard language exists among multiple input languages), as well as the number of input language attempts and the number of keyboard language attempts. This solution effectively improves the accuracy of recommended language updates by accurately determining the backup language and its corresponding backup language score based on the correspondence between the keyboard language and the input language, as well as the number of input language attempts and the number of keyboard language attempts, when the number of keyboard languages ​​is one and the keyboard language is a valid language.

[0091] In one possible embodiment, the recommended language updating method provided by this solution, when determining the backup language and the backup language score corresponding to the backup language based on the correspondence between the keyboard language and the input language, as well as the number of input language times and the number of keyboard language times, includes:

[0092] S2431: If a keyboard language exists in the input languages, the keyboard language is determined as a backup language, and a backup language score corresponding to the backup language is determined based on the sum of the keyboard language usage count of the keyboard language and the input language usage count of the corresponding input language.

[0093] S2432: If the keyboard language does not exist in the input language, determine the keyboard language as a backup language, and determine a backup language score corresponding to the backup language according to the keyboard language count of the keyboard language.

[0094] Exemplarily, when the number of keyboard languages ​​is 1 and the keyboard language is a valid language, it is determined whether the keyboard language corresponds to the input language, that is, whether the keyboard language exists in all input languages.

[0095] In one embodiment, when a keyboard language exists in the input languages, the keyboard language is determined as a standby language, and the sum of the keyboard language count of the keyboard language and the input language count of the corresponding input language is used as the standby language score corresponding to the standby language.

[0096] In one embodiment, when the keyboard language does not exist in the input languages, the keyboard language is determined as a standby language, and the number of keyboard language usages of the keyboard language is used as the standby language score corresponding to the standby language.

[0097] This solution effectively improves the accuracy of recommended language updates by accurately determining the backup language and the backup language score based on the correspondence between the keyboard language and the input language, the number of input languages, and the number of keyboard languages ​​when the number of keyboard languages ​​is 1 and the keyboard language is a valid language.

[0098] S244: When the number of keyboard languages ​​is 1 and the keyboard language is an invalid language, a backup language and a backup language score corresponding to the backup language are determined from the invalid languages ​​according to the correspondence between the keyboard language and the input language, the number of input languages ​​and the number of keyboard languages.

[0099] For example, when the number of keyboard languages ​​is 1 and the keyboard language is an invalid language (e.g., English), the backup language and its corresponding backup language score are determined based on the correspondence between the keyboard language and the input language (i.e., determining whether English is present among the multiple input languages), as well as the number of language inputs and the number of keyboard language attempts. This solution effectively improves the accuracy of recommended language updates by accurately determining the backup language and its corresponding backup language score based on the correspondence between the keyboard language and the input language, as well as the number of language inputs and the number of keyboard language attempts, when the number of keyboard languages ​​is 1 and the keyboard language is an invalid language.

[0100] In one embodiment, the method for updating recommended languages ​​provided by this solution includes determining a backup language and a corresponding backup language score from among the set invalid languages ​​based on the correspondence between the keyboard language and the input language, the number of input languages, and the number of keyboard language attempts, including:

[0101] S2441: When the keyboard language count is greater than the third set count, an invalid language exists among the input languages, and the input language count of the corresponding input language is greater than the fourth set count, the keyboard language is determined as a backup language, and a backup language score corresponding to the backup language is determined based on the sum of the keyboard language count of the keyboard language and the input language count of the corresponding input language.

[0102] S2442: When the number of keyboard language usage times is greater than a fifth set number, the keyboard language is determined as a backup language, and a backup language score corresponding to the backup language is determined according to the number of keyboard language usage times of the keyboard language.

[0103] Exemplarily, when the number of keyboard languages ​​is 1 and the keyboard language is an invalid language (e.g., English), it is determined whether the number of keyboard language uses is greater than a third set number (e.g., the third set number is 3), whether there is an invalid language among the input languages, and whether the number of input languages ​​of the corresponding input language (the input language consistent with the invalid language) is greater than a fourth set number (the fourth set number is less than the third set number, e.g., the fourth set number is 2).

[0104] In one embodiment, when the keyboard language count is greater than a third set count, an invalid language exists among the input languages, and the input language count of the corresponding input language is greater than a fourth set count, the keyboard language (in this case, the invalid language) is determined as a backup language, and the sum of the keyboard language count of the keyboard language and the input language count of the corresponding input language is used as the backup language score corresponding to the backup language.

[0105] In one embodiment, when the keyboard language count is less than or equal to the third set count, there is no invalid language among the input languages, or the input language count of the corresponding input language is less than or equal to the fourth set count, it is determined whether the keyboard language count is greater than the fifth set count (the fifth set count is greater than the third set count, for example, the fifth set count is 10). If the keyboard language count is greater than the fifth set count, the keyboard language (in this case, the invalid language) is determined as a backup language, and the keyboard language count of the keyboard language is used as the backup language score corresponding to the backup language.

[0106] This solution effectively improves the accuracy of recommended language updates by setting the invalid language as the default language when the number of keyboard languages ​​is 1 and the keyboard language is set to an invalid language, and accurately determining the backup language score based on the correspondence between the keyboard language and the input language, as well as the number of input languages ​​and the number of keyboard languages.

[0107] S250: Update the recommended languages ​​according to the current number of recommended languages, as well as the backup languages ​​and the backup language scores.

[0108] Exemplarily, after determining the backup languages ​​and backup language scores, the number of currently recommended languages ​​is determined, and the recommended languages ​​are updated based on the number of recommended languages, the backup languages, and the backup language scores. For example, when the number of recommended languages ​​is 0 or the number of recommended languages ​​does not reach the set number of languages ​​(e.g., 2), the backup language is directly added as a new recommended language, and the recommended language score corresponding to the recommended language is consistent with the backup language score. When the number of recommended languages ​​reaches the set number of languages, the backup language score can be compared with the minimum recommended language score, and based on the comparison result, it is determined whether to replace the recommended language with the minimum recommended language score with the backup language. For example, when the backup language score is greater than the minimum recommended language score, the recommended language score is less than the set score threshold, and / or the backup language score is greater than the set score threshold, it is determined that the recommended language with the minimum recommended language score is replaced with the backup language; otherwise, the current recommended language is retained.

[0109] This solution updates the recommended languages ​​based on the current number of recommended languages, backup languages, and backup language scores to obtain recommended languages ​​that better match the user's language preferences. This can recommend more personalized multilingual content to users, helping the system better select and recommend video content that meets user expectations and provide a richer viewing experience.

[0110] In one possible embodiment, as shown in a flowchart of a recommended language update provided in FIG4 , the recommended language update method provided in this solution updates the recommended language based on the current number of recommended languages, backup languages, and backup language scores, including:

[0111] S251: When the number of currently recommended languages ​​is 1 and there are backup languages ​​other than the recommended languages, the recommended languages ​​are updated based on the recommended language scores corresponding to the recommended languages, the backup languages, and the backup language scores.

[0112] For example, the number of currently recommended languages ​​is determined, and whether there are any non-recommended backup languages ​​(i.e., whether there are any backup languages ​​among the recommended languages) is determined. If the current number of recommended languages ​​is one and there is a non-recommended backup language, the recommended languages ​​are updated based on the recommended language score, the backup language, and the backup language score corresponding to the recommended language. In this case, the number of recommended languages ​​is updated to two.

[0113] For example, when the recommended language score is small (less than the set score threshold) or the recommended language is an invalid language, a backup language is added as a new recommended language to improve the compatibility between the recommended language and the user's language preference. When the recommended language score reaches the set score threshold but the backup language score is small, there is no need to add a new recommended language, which reduces the situation where the recommended language deviates from the user's language preference and improves the accuracy of information pushed to users.

[0114] This solution improves the compatibility of recommended languages ​​with users' language preferences and effectively improves the accuracy of information pushed to users by adding and updating recommended languages ​​based on the recommended language scores, backup languages, and backup language scores corresponding to the recommended languages ​​when the current number of recommended languages ​​is 1 and there are backup languages ​​that are not recommended.

[0115] In one possible embodiment, the recommended language updating method provided by this solution includes:

[0116] S2511: If the recommended language score corresponding to the recommended language is greater than or equal to the first set score and the recommended language is an invalid language, a backup language is added as a recommended language, and the recommended language score of the newly added recommended language is determined based on the backup language score of the backup language.

[0117] S2512: If the recommended language score corresponding to the recommended language is greater than or equal to the first set score, the recommended language is not an invalid language, and the backup language score of the backup language is greater than the first set score, the backup language is added as a recommended language, and the recommended language score of the newly added recommended language is determined based on the backup language score of the backup language.

[0118] S2513: When the recommended language score corresponding to the recommended language is less than the first set score, a new standby language is added as a recommended language, and the recommended language score of the new recommended language is determined according to the standby language score of the standby language.

[0119] Exemplarily, when the recommended language score corresponding to the recommended language is greater than or equal to a first set score (for example, 3 points), and the recommended language is an invalid language (for example, there is a recommended language corresponding to English), a new backup language is added as the recommended language, and the backup language score of the backup language is used as the recommended language score of the newly added recommended language.

[0120] In one embodiment, when the recommended language score corresponding to the recommended language is greater than or equal to a first set score, the recommended language is not an invalid language (for example, there is no recommended language corresponding to English), and the standby language score of the standby language is greater than the first set score, the standby language is added as a recommended language, and the standby language score of the standby language is used as the recommended language score of the newly added recommended language.

[0121] In one embodiment, when the recommended language score corresponding to the recommended language is less than a first set score, the standby language is used as a new recommended language, and the standby language score of the standby language is used as the recommended language score of the new recommended language.

[0122] This solution accurately updates the recommended languages ​​based on the recommended language score, backup language, and backup language score corresponding to the recommended language when the current number of recommended languages ​​is 1 and the backup language is not among the recommended languages. This improves the compatibility of the recommended language with the user's language preference and effectively improves the accuracy of information pushed to users.

[0123] S252: When the number of currently recommended languages ​​is greater than 1, the recommended languages ​​are updated based on the recommended language scores corresponding to one or more second recommended languages ​​ranked later, as well as the backup languages ​​and their backup language scores.

[0124] This solution defines one or more recommended languages ​​ranked at the end of the recommended language scores among multiple recommended languages ​​as the second recommended language (for example, the recommended language with the smallest recommended language score), and the recommended language ranked first in recommended language score is defined as the first recommended language. Exemplarily, when the number of currently recommended languages ​​is greater than 1 (for example, there are two recommended languages), the recommended languages ​​are replaced and updated based on the recommended language scores corresponding to the one or more second recommended languages ​​ranked at the end, as well as the backup languages ​​and their scores. For example, when the second recommended language score is smaller, or the backup language score is greater than the recommended language score of the second recommended language, the second recommended language is replaced with the backup language. When the backup language score is smaller, the current recommended language combination can be maintained to reduce the situation where the recommended language deviates from the user's language preference and improve the accuracy of information pushed to the user.

[0125] In one possible embodiment, the recommended language updating method provided by this solution, when performing language replacement and updating of the recommended language based on the recommended language scores corresponding to one or more second recommended languages ​​ranked later, as well as the backup languages ​​and their backup language scores, includes:

[0126] S2521: When the recommended language scores corresponding to one or more second recommended languages ​​ranked later are less than or equal to the second set score, the second recommended language is replaced with a backup language, and the recommended language score of the newly added recommended language is determined based on the backup language scores of the backup languages.

[0127] S2522: When the recommended language scores corresponding to one or more second recommended languages ​​ranked later are greater than the third set score, and the backup language score of the backup language is greater than the fourth set score, the second recommended language is replaced with the backup language, and the recommended language score of the newly added recommended language is determined based on the backup language score of the backup language.

[0128] For example, when the recommended language scores corresponding to one or more second recommended languages ​​ranked later are less than or equal to a second set score (the first set score may be the same as the second set score, for example, the second set score is 3), the second recommended language is replaced with a backup language, and the backup language score of the backup language is used as the recommended language score of the newly added recommended language. When the recommended language scores corresponding to one or more second recommended languages ​​ranked later are greater than a third set score (the third set score is greater than the second set score, for example, the third set score is 3), and the backup language score of the backup language is greater than a fourth set score (the fourth set score is greater than the third set score, for example, the fourth set score is 5), the second recommended language is replaced with the backup language, and the backup language score of the backup language is used as the recommended language score of the newly added recommended language (in this case, the original second recommended language is already relatively commonly used, and the backup language score of the backup language has significantly improved compared to the recommended language score of the second recommended language, so the backup language is replaced). This solution replaces and updates the recommended language based on the recommended language score, backup language, and backup language score corresponding to the second recommended language when the current number of recommended languages ​​is greater than 1, thereby improving the compatibility of the recommended language with the user's language preference and effectively improving the accuracy of information pushed to users.

[0129] In one possible embodiment, after updating the recommended languages ​​based on the backup languages ​​and their scores, the recommended language updating method provided by this solution further includes: determining the proportion of recommended content corresponding to each recommended language based on the recommended language scores corresponding to each recommended language, where the recommended content proportion is used to determine the language proportion of content recommended to the user.

[0130] Exemplarily, the proportion of the recommended language score of each recommended language in the total recommended language score is determined according to the recommended language score corresponding to each recommended language, and the proportion of the recommended language score corresponding to each recommended language is used as the recommended content proportion of the corresponding recommended language.

[0131] When sending recommended content to a user, the language ratio of the recommended content can be determined based on the recommended content ratio. This allows the recommended language ratio to better align with the user's language preference distribution, improving the user experience. For example, for the recommended language score set {ar:8,bn:2}, where the recommended language scores for the first recommended language (Arabic ar) and the second recommended language (Bengali bn) are 8 and 2, respectively, then the first and second recommended languages ​​account for 8 / (8+2)=80% and 2 / (8+2)=20% of the total recommended language score, respectively. This means that the recommended content ratios for the first and second recommended languages ​​are 80% and 20%, respectively. The content recommended to the user in Arabic and Bengali has a ratio of 80%:20%. This solution flexibly adjusts the language ratio of content recommended to users by determining the recommended content ratio for each language based on the recommended language score. This allows the recommended language ratio to better align with the user's language preference distribution, improving the user experience.

[0132] As described above, by determining one or more input languages ​​and the number of input language uses corresponding to each input language based on user input data, and determining one or more keyboard languages ​​and the number of keyboard language uses corresponding to each keyboard language based on keyboard usage data, determining backup languages ​​and backup language scores corresponding to the backup languages ​​based on the input language, the number of input language uses, the keyboard language, and the number of keyboard language uses, and updating the recommended languages ​​based on the backup languages ​​and the backup language scores, and by determining backup languages ​​and the corresponding backup language scores based on user input data and keyboard usage data that reflect user behavior, the recommended languages ​​can be dynamically corrected, and flexible updates of the recommended languages ​​can be achieved. The recommended languages ​​are more closely aligned with user behavior habits, better meet user needs for multilingual content, and improve information recommendation effects. At the same time, by accurately determining the backup languages ​​and their corresponding backup language scores based on the input language, number of inputs, keyboard language, number of keyboard language inputs, and the validity of the keyboard languages, the accuracy of recommended language updates is improved, helping the system to more accurately correct and adjust the user's language preferences, thereby providing more appropriate multilingual content recommendations, making users more satisfied with the app's content recommendations and experience, and effectively improving user satisfaction. Furthermore, by updating the recommended languages ​​based on the current number of recommended languages, backup languages, and backup language scores, recommended languages ​​that better match the user's language preferences are obtained, allowing for more personalized multilingual content recommendations for users. This helps the system better select and recommend video content that meets user expectations, providing a richer viewing experience.

[0133] FIG5 is a schematic diagram of the structure of a recommended language updating device provided by an embodiment of the present application. Referring to FIG5 , the recommended language updating device includes an input language module 51 , a keyboard language module 52 , a backup language module 53 and a backup language module 54 .

[0134] Among them, the input language module 51 is configured to determine one or more input languages ​​and the number of input language uses corresponding to each input language based on user input data; the keyboard language module 52 is configured to determine one or more keyboard languages ​​and the number of keyboard language uses corresponding to each keyboard language based on keyboard usage data; the backup language module 53 is configured to determine the backup language and the backup language score corresponding to the backup language based on the input language, the number of input language uses, the keyboard language, and the number of keyboard language uses; the language update module 54 is configured to update the recommended language based on the backup language and the backup language score.

[0135] As described above, by determining one or more input languages ​​and the number of input language uses corresponding to each input language based on user input data, and determining one or more keyboard languages ​​and the number of keyboard language uses corresponding to each keyboard language based on keyboard usage data, determining backup languages ​​and backup language scores corresponding to the backup languages ​​based on the input language, the number of input language uses, the keyboard language, and the number of keyboard language uses, and updating the recommended languages ​​based on the backup languages ​​and the backup language scores, and by determining backup languages ​​and the corresponding backup language scores based on user input data and keyboard usage data that reflect user behavior, the recommended languages ​​can be dynamically corrected, and flexible updates of the recommended languages ​​can be achieved. The recommended languages ​​are more closely aligned with user behavior habits, better meet user needs for multilingual content, and improve information recommendation effects.

[0136] In a possible embodiment, when the input language module 51 determines one or more input languages ​​and the number of input times corresponding to each input language based on user input data, it is configured as follows:

[0137] Obtaining user input data and performing text preprocessing on the user input data to obtain user input text data, where the user input data includes one or more combinations of description data, comment data, nickname data, signature data, and chat data, and the text preprocessing includes one or more combinations of punctuation removal, emoticon removal, normalization, and full-width / half-width conversion;

[0138] Perform language recognition on the text data input by the user to obtain one or more input languages, and determine the number of input languages ​​corresponding to each input language.

[0139] In a possible embodiment, when the backup language module 53 determines the backup language and the backup language score corresponding to the backup language based on the input language, the number of language inputs, the keyboard language, and the number of keyboard inputs, it is configured as follows:

[0140] The backup language and the backup language score corresponding to the backup language are determined based on the number of keyboard languages ​​and the validity of the keyboard languages, as well as the input language, the number of input times, the keyboard language and the number of keyboard language times.

[0141] In one possible embodiment, the backup language module 53 is configured to determine the backup language and the corresponding backup language score based on the number of keyboard languages ​​and the validity of the keyboard languages, as well as the input language, the number of language inputs, the keyboard language, and the number of keyboard language inputs:

[0142] If the number of keyboard languages ​​is greater than one and the first keyboard language with the largest number of keyboard language uses is a valid language, determine a backup language and a backup language score corresponding to the backup language based on the correspondence between the first keyboard language and the input language, the number of input language uses, and the number of keyboard language uses;

[0143] If the number of keyboard languages ​​is greater than one, the first keyboard language with the most keyboard language usage is an invalid language, and the second keyboard language with the second most keyboard language usage is a valid language, a backup language and a backup language score corresponding to the backup language are determined based on the correspondence between the second keyboard language and the input language, as well as the number of input language usages and the number of keyboard language usages.

[0144] When the number of keyboard languages ​​is 1 and the keyboard language is a valid language, determine the backup language and the backup language score corresponding to the backup language based on the correspondence between the keyboard language and the input language, as well as the number of input languages ​​and the number of keyboard languages ​​used;

[0145] When the number of keyboard languages ​​is 1 and the keyboard language is an invalid language, a backup language and a backup language score corresponding to the backup language are determined from the invalid languages ​​according to the correspondence between the keyboard language and the input language, the number of input languages ​​and the number of keyboard languages.

[0146] In one possible embodiment, when the backup language module 53 determines the backup language and the backup language score corresponding to the backup language based on the correspondence between the first keyboard language and the input language, the number of input language attempts, and the number of keyboard language attempts, it is configured as follows:

[0147] If the first keyboard language exists in the input languages, the first keyboard language is determined as a backup language, and a backup language score corresponding to the backup language is determined based on the sum of the keyboard language count of the first keyboard language and the input language count of the corresponding input language;

[0148] When the first keyboard language does not exist in the input languages, the input language with the largest number of input times is determined as the standby language, and a standby language score corresponding to the standby language is determined according to the number of input times of the input language.

[0149] In one possible embodiment, when the backup language module 53 determines the backup language and the backup language score corresponding to the backup language based on the correspondence between the second keyboard language and the input language, as well as the number of input language attempts and the number of keyboard language attempts, it is configured as follows:

[0150] If a second keyboard language exists among the input languages, the second keyboard language is determined as a backup language, and a backup language score corresponding to the backup language is determined based on the sum of the keyboard language count of the second keyboard language and the input language count of the corresponding input language;

[0151] If the second keyboard language does not exist in the input languages ​​and the keyboard language count of the second keyboard language is greater than the first set count, the second keyboard language is determined as a backup language, and a backup language score corresponding to the backup language is determined based on the keyboard language count of the second keyboard language;

[0152] If the second keyboard language does not exist in the input languages, the keyboard language count of the second keyboard language is less than or equal to the first set count, and the first keyboard language is an invalid language and the keyboard language count is greater than the second set count, the first keyboard language is determined as a backup language, and a backup language score corresponding to the backup language is determined based on the keyboard language count of the first keyboard language;

[0153] If the second keyboard language does not exist in the input languages, the keyboard language count of the second keyboard language is less than or equal to the first set count, and the first keyboard language is a non-set invalid language and / or the keyboard language count of the first keyboard language is less than or equal to the second set count, the second keyboard language is determined as a backup language, and a backup language score corresponding to the backup language is determined based on the keyboard language count of the second keyboard language.

[0154] In a possible embodiment, the backup language module 53 determines the backup language and the backup language score corresponding to the backup language based on the correspondence between the keyboard language and the input language, the number of input language attempts, and the number of keyboard language attempts, including:

[0155] If a keyboard language exists among the input languages, the keyboard language is determined as a backup language, and a backup language score corresponding to the backup language is determined based on the sum of the keyboard language usage count of the keyboard language and the input language usage count of the corresponding input language;

[0156] In the case that the keyboard language does not exist in the input language, the keyboard language is determined as the backup language, and the backup language score corresponding to the backup language is determined according to the keyboard language count of the keyboard language.

[0157] In one possible embodiment, the backup language module 53 is configured to determine the backup language and the corresponding backup language score from the set invalid languages ​​based on the correspondence between the keyboard language and the input language, the number of input language attempts, and the number of keyboard language attempts:

[0158] If the keyboard language usage count is greater than a third set count, an invalid language is included in the input languages, and the input language usage count of the corresponding input language is greater than a fourth set count, the keyboard language is determined as a backup language, and a backup language score corresponding to the backup language is determined based on the sum of the keyboard language usage count of the keyboard language and the input language usage count of the corresponding input language;

[0159] When the number of keyboard language usage times is greater than a fifth set number, the keyboard language is determined as a standby language, and a standby language score corresponding to the standby language is determined according to the number of keyboard language usage times of the keyboard language.

[0160] In one possible embodiment, when the language updating module 54 updates the recommended language based on the backup languages ​​and the backup language scores, it is configured as follows:

[0161] The recommended languages ​​are updated based on the current number of recommended languages, as well as the backup languages ​​and their scores.

[0162] In one possible embodiment, when the language updating module 54 updates the recommended languages ​​based on the current number of recommended languages, the backup languages, and the backup language scores, it is configured to:

[0163] If the number of currently recommended languages ​​is 1 and there are backup languages ​​other than the recommended languages, the recommended languages ​​are updated based on the recommended language scores corresponding to the recommended languages, as well as the backup languages ​​and their backup language scores.

[0164] When the number of currently recommended languages ​​is greater than 1, the recommended languages ​​are replaced and updated according to the recommended language scores corresponding to one or more second recommended languages ​​ranked later, as well as the backup languages ​​and their backup language scores.

[0165] In one possible embodiment, when the language updating module 54 adds and updates the recommended languages ​​based on the recommended language scores corresponding to the recommended languages, as well as the backup languages ​​and their backup language scores, it is configured as follows:

[0166] If the recommended language score corresponding to the recommended language is greater than or equal to the first set score and the recommended language is an invalid language, a backup language is added as the recommended language, and the recommended language score of the newly added recommended language is determined based on the backup language score of the backup language;

[0167] If the recommended language score corresponding to the recommended language is greater than or equal to the first set score, the recommended language is not an invalid language, and the backup language score of the backup language is greater than the first set score, the backup language is added as a recommended language, and the recommended language score of the newly added recommended language is determined based on the backup language score of the backup language;

[0168] When the recommended language score corresponding to the recommended language is less than the first set score, a new standby language is added as the recommended language, and the recommended language score of the new recommended language is determined according to the standby language score of the standby language.

[0169] In one possible embodiment, when the language updating module 54 performs language replacement and updating based on the recommended language scores corresponding to one or more second recommended languages ​​ranked later, as well as the backup languages ​​and their backup language scores, the configuration is as follows:

[0170] If the recommended language scores corresponding to one or more second recommended languages ​​ranked later are less than or equal to the second set score, replacing the second recommended language with the backup language, and determining the recommended language score of the newly added recommended language based on the backup language scores of the backup languages;

[0171] If the recommended language scores corresponding to one or more second recommended languages ​​ranked later are greater than the third set score, and the backup language score of the backup language is greater than the fourth set score, the second recommended language is replaced by the backup language, and the recommended language score of the newly added recommended language is determined based on the backup language score of the backup language.

[0172] In a possible embodiment, the recommended language updating device further includes a ratio determination module, which is configured to determine the ratio of recommended content corresponding to each recommended language based on the recommended language score corresponding to each recommended language. The recommended content ratio is used to determine the language ratio of content recommended to the user.

[0173] It is worth noting that in the embodiment of the above-mentioned recommended language updating device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the embodiments of this application.

[0174] The embodiment of the present application also provides a recommended language update device, which can integrate the recommended language update apparatus provided in the embodiment of the present application. Figure 6 is a structural diagram of a recommended language update device provided in the embodiment of the present application. Referring to Figure 6, the recommended language update device includes: an input device 63, an output device 64, a memory 62 and one or more processors 61; the memory 62 is used to store one or more programs; when the one or more programs are executed by one or more processors 61, the one or more processors 61 implement the recommended language update method provided in the above embodiment. The recommended language update apparatus, device and computer provided above can be used to execute the recommended language update method provided in any of the above embodiments, and have corresponding functions and beneficial effects.

[0175] The embodiments of the present application also provide a non-volatile storage medium storing computer-executable instructions, which, when executed by a computer processor, are used to execute the recommended language update method provided in the above embodiments. Of course, the non-volatile storage medium storing computer-executable instructions provided in the embodiments of the present application, whose computer-executable instructions are not limited to the recommended language update method provided above, can also execute the related operations in the recommended language update method provided in any embodiment of the present application. The recommended language update apparatus, device and storage medium provided in the above embodiments can execute the recommended language update method provided in any embodiment of the present application. For technical details not fully described in the above embodiments, please refer to the recommended language update method provided in any embodiment of the present application.

[0176] Based on the above embodiments, the embodiments of the present application also provide a computer program product. The essence of the technical solution of the present application or the part that contributes to the existing technology or all or part of the technical solution can be embodied in the form of a software product. The computer program product is stored in a storage medium and includes a number of instructions for enabling a computer device, a mobile terminal or a processor therein to execute all or part of the steps of the recommended language update method provided in each embodiment of the present application.

Claims

1. A method for updating a recommended language, wherein: include: Determine one or more input languages ​​and the number of input languages ​​corresponding to each of the input languages ​​according to the user input data; Determine one or more keyboard languages ​​and the number of keyboard language usages corresponding to each of the keyboard languages ​​according to the keyboard usage data; Determine a backup language and a backup language score corresponding to the backup language according to the input language, the number of times the language is input, the keyboard language and the number of times the keyboard language is input; The recommended language is updated according to the backup language and the backup language score.

2. The method for updating the recommended language according to claim 1, wherein: The step of determining one or more input languages ​​and the number of input languages ​​corresponding to each of the input languages ​​according to the user input data includes: Acquire user input data, and perform text preprocessing on the user input data to obtain user input text data, wherein the user input data includes one or more combinations of description data, comment data, nickname data, signature data, and chat data, and the text preprocessing includes one or more combinations of punctuation removal processing, emoticon removal processing, normalization processing, and full-width and half-width conversion processing; Perform language recognition on the user input text data to obtain one or more input languages, and determine the number of input languages ​​corresponding to each of the input languages.

3. The method for updating the recommended language according to claim 1, wherein: The step of determining the standby language and the standby language score corresponding to the standby language according to the input language, the number of times the language is input, the keyboard language and the number of times the keyboard language is input includes: A standby language and a standby language score corresponding to the standby language are determined according to the number of the keyboard languages ​​and the validity of the keyboard languages, as well as the input language, the number of input language times, the keyboard language and the number of keyboard language times.

4. The method for updating the recommended language according to claim 3, wherein: The step of determining the standby language and the standby language score corresponding to the standby language according to the number of the keyboard languages ​​and the validity of the keyboard languages, the input language, the number of input language times, the keyboard language and the number of keyboard language times includes: When the number of the keyboard languages ​​is greater than 1 and the first keyboard language with the largest number of keyboard language usages is a valid language, determining a backup language and a backup language score corresponding to the backup language according to a correspondence between the first keyboard language and the input language, and the number of input language usages and the number of keyboard language usages; When the number of the keyboard languages ​​is greater than 1, the first keyboard language with the largest number of keyboard languages ​​is an invalid language, and the second keyboard language with the second largest number of keyboard languages ​​is a valid language, determining a backup language and a backup language score corresponding to the backup language according to a correspondence between the second keyboard language and the input language, and the number of input languages ​​and the number of keyboard languages; When the number of the keyboard languages ​​is 1 and the keyboard language is a valid language, determining a standby language and a standby language score corresponding to the standby language according to the correspondence between the keyboard language and the input language, the input language count and the keyboard language count; When the number of keyboard languages ​​is 1 and the keyboard language is an invalid language, a backup language and a backup language score corresponding to the backup language are determined from the invalid languages ​​according to the corresponding relationship between the keyboard language and the input language, the input language count and the keyboard language count.

5. The method for updating the recommended language according to claim 4, wherein: The determining of the standby language and the standby language score corresponding to the standby language according to the correspondence between the first keyboard language and the input language, the input language times and the keyboard language times includes: If the first keyboard language exists in the input languages, determining the first keyboard language as a standby language, and determining a standby language score corresponding to the standby language according to the sum of the keyboard language count of the first keyboard language and the input language count of the corresponding input language; When the first keyboard language does not exist in the input languages, the input language with the largest number of input times is determined as a standby language, and a standby language score corresponding to the standby language is determined according to the number of input times of the input language.

6. The method for updating the recommended language according to claim 4, wherein: The determining of the standby language and the standby language score corresponding to the standby language according to the correspondence between the second keyboard language and the input language, the input language times and the keyboard language times includes: If the second keyboard language exists in the input languages, determining the second keyboard language as a standby language, and determining a standby language score corresponding to the standby language according to the sum of the keyboard language times of the second keyboard language and the input language times of the corresponding input language; If the second keyboard language does not exist in the input languages ​​and the keyboard language number of the second keyboard language is greater than the first set number, the second keyboard language is determined as a standby language, and a standby language score corresponding to the standby language is determined according to the keyboard language number of the second keyboard language; If the second keyboard language does not exist in the input languages, the keyboard language number of the second keyboard language is less than or equal to the first set number, and the first keyboard language is an invalid language and the keyboard language number is greater than the second set number, the first keyboard language is determined as a standby language, and a standby language score corresponding to the standby language is determined according to the keyboard language number of the first keyboard language; If the second keyboard language does not exist in the input languages, the keyboard language number of the second keyboard language is less than or equal to the first set number, and the first keyboard language is a non-set invalid language and / or the keyboard language number of the first keyboard language is less than or equal to the second set number, the second keyboard language is determined as a standby language, and the standby language score corresponding to the standby language is determined according to the keyboard language number of the second keyboard language.

7. The method for updating the recommended language according to claim 4, wherein: The determining of the standby language and the standby language score corresponding to the standby language according to the correspondence between the keyboard language and the input language, the input language times and the keyboard language times includes: In the case where the keyboard language exists in the input language, determining the keyboard language as a standby language, and determining a standby language score corresponding to the standby language according to the sum of the keyboard language times of the keyboard language and the input language times of the corresponding input language; In the case that the keyboard language does not exist in the input language, the keyboard language is determined as a standby language, and a standby language score corresponding to the standby language is determined according to the keyboard language times of the keyboard language.

8. The method for updating recommended languages ​​according to claim 4, wherein: The determining of a standby language and a standby language score corresponding to the standby language from the set invalid languages ​​according to the correspondence between the keyboard language and the input language, the input language times and the keyboard language times, comprises: When the keyboard language number is greater than the third set number, the set invalid language exists among the input languages, and the input language number of the corresponding input language is greater than the fourth set number, the keyboard language is determined as a standby language, and a standby language score corresponding to the standby language is determined according to the sum of the keyboard language number of the keyboard language and the input language number of the corresponding input language; When the keyboard language number is greater than a fifth set number, the keyboard language is determined as a standby language, and a standby language score corresponding to the standby language is determined according to the keyboard language number of the keyboard language.

9. The method for updating recommended languages ​​according to claim 1, wherein: The updating of the recommended language according to the standby language and the standby language score includes: The recommended languages ​​are updated according to the number of current recommended languages, the backup languages, and the backup language scores.

10. The method for updating the recommended language according to claim 9, wherein: The updating of the recommended languages ​​according to the number of current recommended languages, the backup languages, and the backup language scores includes: When the number of currently recommended languages ​​is 1 and there is a backup language that is not a recommended language, the recommended language is newly added and updated according to the recommended language score corresponding to the recommended language, the backup language, and the backup language score; When the number of currently recommended languages ​​is greater than 1, the recommended language is updated by replacing the recommended language according to the recommended language scores corresponding to one or more second recommended languages ​​ranked later, the backup language, and the backup language scores.

11. The method for updating recommended languages ​​according to claim 10, wherein: The updating of the recommended language according to the recommended language score corresponding to the recommended language, the backup language and the backup language score includes: When the recommended language score corresponding to the recommended language is greater than or equal to the first set score and the recommended language is a set invalid language, adding the standby language as a recommended language, and determining the recommended language score of the newly added recommended language according to the standby language score of the standby language; If the recommended language score corresponding to the recommended language is greater than or equal to the first set score, the recommended language is not a set invalid language, and the standby language score of the standby language is greater than the first set score, the standby language is added as a recommended language, and the recommended language score of the newly added recommended language is determined according to the standby language score of the standby language; When the recommended language score corresponding to the recommended language is less than the first set score, the standby language is added as a recommended language, and the recommended language score of the newly added recommended language is determined according to the standby language score of the standby language.

12. The method for updating recommended languages ​​according to claim 10, wherein: The updating of the recommended language according to the recommended language scores corresponding to the one or more second recommended languages ​​ranked later, the backup language and the backup language scores, includes: When the recommended language scores corresponding to the one or more second recommended languages ​​ranked later are less than or equal to the second set score, replacing the second recommended language with the backup language, and determining the recommended language score of the newly added recommended language according to the backup language scores of the backup languages; When the recommended language scores corresponding to one or more second recommended languages ​​ranked later are greater than the third set score, and the standby language score of the standby language is greater than the fourth set score, the second recommended language is replaced by the standby language, and the recommended language score of the newly added recommended language is determined according to the standby language score of the standby language.

13. The method for updating recommended languages ​​according to any one of claims 1 to 12, wherein: After the recommended language is updated according to the standby language and the standby language score, the method further includes: According to the recommended language scores corresponding to the recommended languages, the recommended content ratios corresponding to the recommended languages ​​are determined, and the recommended content ratios are used to determine the language ratios of the content recommended to the user.

14. A device for updating recommended languages, wherein: It includes input language module, keyboard language module, backup language module and backup language module, among which: The input language module is configured to determine one or more input languages ​​and the number of input languages ​​corresponding to each of the input languages ​​according to the user input data; The keyboard language module is configured to determine one or more keyboard languages ​​and the number of keyboard language times corresponding to each of the keyboard languages ​​according to the keyboard usage data; The standby language module is configured to determine a standby language and a standby language score corresponding to the standby language according to the input language, the number of input language times, the keyboard language and the number of keyboard language times; The language updating module is configured to update the recommended language according to the standby language and the standby language score.

15. A device for updating recommended languages, wherein: include: memory and one or more processors; The memory is used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the recommended language updating method according to any one of claims 1 to 13.

16. A non-volatile storage medium storing computer executable instructions, wherein: The computer executable instructions are used to execute the recommended language updating method according to any one of claims 1 to 13 when executed by a computer processor.

17. A computer program product comprising a computer program, wherein: When the computer program is executed by a processor, the method for updating the recommended language according to any one of claims 1 to 13 is implemented.

Images