Input method word bank generation method, text input method, related device and medium

Abstract

The invention relates to the field of computer input methods, in particular to an input method lexicon generation method, a text input method, related devices and a medium. The method for generating the word bank of the input method comprises the following steps of: firstly, acquiring an object feature reference of a target group, acquiring an object feature of each object in an object set using the input method, acquiring objects of which the object features are matched with the object feature reference in the object set to form the target group, and generating the word bank of the input method on the basis of a word input historical record of each object in the target group. And generating a group lexicon of the target group for the input method. On the other hand, according to the text input method, a first code string input by a target object needs to be received in an input method interface, and then a group lexicon control matched with the first code string is displayed. And further, in response to triggering of the group lexicon control, obtaining the group lexicon so as to generate a target text and display the target text on the input method interface. The text input efficiency of the input method can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of computer input methods, and in particular to an input method word library generation method, a text input method, and related devices and media. Background Art

[0002] Current input methods generally use a general word library. When generating a general word library, the frequency of input words of all input method objects is counted, and the number of times each input word is input by all input method objects is counted. The input words with the top input times are selected as the general word library. When an object uses the input method to input, the general word library is searched according to the code string (pinyin, etc.) input by the object, and the target text corresponding to the code string retrieved in the general word library is displayed for the object to select.

[0003] Today's society is developing rapidly, and people often have different text input demands. When the word that the object inputs has a low frequency in the whole domain, but is a word that the object in a certain field or range often uses, it may be difficult to find the word in the general word library. In this case, the object needs to search and combine words one by one in the input method interface, resulting in low input efficiency. Summary of the invention

[0004] The embodiments of the present disclosure provide an input method word library generation method, a text input method, related devices and media, which can improve the text input efficiency of the input method.

[0005] According to one aspect of the present disclosure, there is provided a method for generating an input method vocabulary, comprising:

[0006] Obtaining object feature benchmarks for the target group;

[0007] Obtaining object features of each object in an object set using the input method;

[0008] In the object set, obtaining objects whose object features match the object feature benchmark to form the target group;

[0009] Based on the word input history records of each of the objects in the target group, a group word library of the target group is generated for use in the input method.

[0010] According to one aspect of the present disclosure, there is provided a text input method, comprising:

[0011] In an input method interface, receiving a first encoding string corresponding to a target text input by a target object, the input method having a universal word library;

[0012] Displaying a group vocabulary control matching the first encoding string, wherein the group vocabulary control corresponds to the target group to which the target object belongs;

[0013] In response to triggering the group vocabulary control, a group vocabulary is obtained so as to convert the second encoding string input by the target object into the target text according to the general vocabulary and the group vocabulary; wherein the group vocabulary is generated by the input method vocabulary generation method described in one aspect of the present disclosure;

[0014] The target text is displayed.

[0015] According to one aspect of the present disclosure, there is provided an input method word library generating device, comprising:

[0016] A first acquisition unit, used to acquire an object feature reference of a target group;

[0017] A second acquisition unit, used to acquire an object feature of each object in an object set using the input method;

[0018] A third acquisition unit, configured to acquire objects in the object set that match the object feature with the object feature reference to form the target group;

[0019] The first generating unit is configured to generate a group word library of the target group based on the word input history records of each of the objects in the target group, for use in the input method.

[0020] According to one aspect of the present disclosure, there is provided a text input device, comprising:

[0021] A first receiving unit, used for receiving a first encoding string corresponding to a target text input by a target object in an input method interface, wherein the input method has a universal word library;

[0022] A first display unit, configured to display a group vocabulary control matching the first coding string, wherein the group vocabulary control corresponds to the target group to which the target object belongs;

[0023] a fourth acquisition unit, configured to acquire a group vocabulary in response to triggering the group vocabulary control, so as to convert the second encoding string input by the target object into the target text according to the general vocabulary and the group vocabulary; wherein the group vocabulary is generated by the input method vocabulary generation method described in one aspect of the present disclosure;

[0024] The second display unit is used to display the target text.

[0025] Optionally, the object feature benchmark includes an object terminal application benchmark, a seed word benchmark and a positioning information benchmark; the object feature includes an application installed on the object terminal of the object, a word input history record of the object, and the positioning information of the object;

[0026] The third acquisition unit is specifically used for:

[0027] For the object in the object set, determining that an application installed on an object terminal of the object meets the object terminal application benchmark;

[0028] Determining that the subject's word input history matches the seed word benchmark;

[0029] Determining that the positioning information of the object matches the positioning information reference;

[0030] The subject is added to the target group.

[0031] Optionally, the seed word benchmark includes a seed word and a first discrimination score corresponding to the seed word;

[0032] The third acquisition unit is specifically used for:

[0033] Set a counter, the counter is initialized to 0;

[0034] Traversing the word input history records of the object, if one of the word input history records corresponds to a seed word, accumulating the first discrimination score corresponding to the seed word into the counter;

[0035] After traversing the word input history record of the object, based on the value of the counter, it is determined that the word input history record of the object meets the seed word benchmark.

[0036] Optionally, the first acquisition unit is specifically used to acquire the seed word, and the seed word is acquired through the following process:

[0037] Acquire a first seed group and a second seed group, wherein the first seed group includes a plurality of first objects that are predetermined to belong to the target group, and the second seed group includes a plurality of second objects that are predetermined to not belong to the target group;

[0038] Acquire a first corpus, where the first corpus includes a plurality of first sample corpora of the first objects;

[0039] Acquire a second corpus, where the second corpus includes a plurality of second sample corpora of the second object;

[0040] Segment each first sample corpus in the first corpus to obtain candidate words;

[0041] For each of the candidate words, determining a first number of occurrences of the candidate word in the first corpus and a second number of occurrences of the candidate word in the second corpus;

[0042] The seed word is determined from the candidate words based on the first number of times and the second number of times.

[0043] Optionally, the first obtaining unit is specifically used to obtain the first discrimination score, and the first discrimination score is obtained by the following process:

[0044] Acquire a third corpus, wherein the third corpus includes sample corpora of a first number of third objects;

[0045] Obtain the third number of times the seed word appears in the third corpus;

[0046] Acquire a second number, wherein in the third corpus, sample corpora of the second number of objects contain the seed word, and the second number is not greater than the first number;

[0047] Based on the third number and the second number, the first discrimination score is determined.

[0048] Optionally, the first acquiring unit is specifically used for:

[0049] Get the first constant;

[0050] The first discrimination score is determined based on the logarithm of the third number, the first constant, and the logarithm of the second number.

[0051] Optionally, the object feature benchmark includes an object terminal application benchmark, a seed word benchmark and a positioning information benchmark; the object feature includes an application installed on the object terminal of the object, a word input history record of the object, and the positioning information of the object;

[0052] The third acquisition unit is specifically used for:

[0053] For the object in the object set, determining a first matching degree between an application installed on an object terminal of the object and an application benchmark of the object terminal;

[0054] determining a second degree of match between the subject's word input history record and the seed word benchmark;

[0055] Determining a third degree of matching between the positioning information of the object and the positioning information reference;

[0056] Determining a degree of match between the object feature and the object feature benchmark based on the first degree of match, the second degree of match, and the third degree of match;

[0057] Based on the matching degree, an object whose object feature matches the object feature reference is obtained.

[0058] Optionally, the seed word benchmark includes a seed word and a first discrimination score corresponding to the seed word;

[0059] The third acquisition unit is specifically used for:

[0060] Set a counter, the counter is initialized to 0;

[0061] Traversing the word input history records of the object, if one of the word input history records corresponds to a seed word, accumulating the first discrimination score corresponding to the seed word into the counter;

[0062] After traversing the word input history of the object, the second matching degree is determined based on the ratio of the value of the counter to a second threshold.

[0063] Optionally, the object feature benchmark includes an object terminal application benchmark, a seed word benchmark and a positioning information benchmark;

[0064] The third acquisition unit is specifically used for:

[0065] For the object terminal application benchmark, the seed word benchmark, and a single object feature benchmark in the positioning information benchmark, obtaining an object whose object feature matches the single object feature benchmark to form a first group corresponding to the single object feature benchmark;

[0066] If the accuracy of one of the first groups is greater than a third threshold, taking the first group as the target group;

[0067] If the accuracy of each of the first groups is not greater than a third threshold, select any two object feature benchmarks from the object terminal application benchmark, the seed word benchmark, and the positioning information benchmark to form an object feature benchmark combination, and for the object feature benchmark combination, obtain objects whose object features match the object feature benchmark combination to form a second group corresponding to the object feature benchmark combination;

[0068] If the accuracy of one of the second groups is greater than the third threshold, taking the second group as the target group;

[0069] If the accuracy of each of the second groups is not greater than the third threshold, objects whose object features match the object terminal application benchmark, the seed word benchmark and the positioning information benchmark are obtained to form the target group.

[0070] Optionally, the third acquiring unit is specifically used for:

[0071] If the accuracy of the plurality of first groups is greater than a third threshold, taking the first group with the greatest accuracy as the target group;

[0072] If the accuracy of a plurality of the second groups is greater than the third threshold, the second group with the greatest accuracy is used as the target group.

[0073] Optionally, the third acquisition unit is specifically used to determine the accuracy of the first group or the second group, and the accuracy of the first group or the second group is determined in the following manner:

[0074] Acquire an evaluation object set, wherein the evaluation object set has a plurality of sample objects, and each of the sample objects has a group label to which it belongs;

[0075] Acquire a fourth object belonging to a plurality of the sample objects from among the objects in the first group or the second group;

[0076] Determine a fifth object whose group label corresponding to the fourth object matches the first group or the second group;

[0077] The ratio of the fifth number of objects to the fourth number of objects is used as the accuracy.

[0078] Optionally, the first generating unit is specifically used for:

[0079] For each word in the word input history, determining a second discrimination score for the word;

[0080] Based on the second discrimination score, a target word is determined from the words in the word input history record and added to the group word library.

[0081] Optionally, the first generating unit is specifically used for:

[0082] For each word in the word input history record, determining a fourth number of times the word appears in the word input history record;

[0083] Acquire a third number, wherein among the word input history records of each of the objects, the word input history records of the third number of objects contain the word;

[0084] Based on the fourth number of times and the third number, a second discrimination score for the word is determined.

[0085] Optionally, the first generating unit is specifically used for:

[0086] For each word in the word input history record, word cleaning is performed;

[0087] For each of the words remaining after word cleaning, a second discrimination score for the word is determined.

[0088] Optionally, the first generating unit is specifically used for:

[0089] Identify the traditional Chinese characters in the words, and delete the words containing the traditional Chinese characters;

[0090] Compare the word with an exclusion word library, and if the word is identical to an exclusion word in the exclusion word library, delete the word;

[0091] The word is compared with a standard word library, and if the word is in the standard word library, the word is retained.

[0092] Optionally, after generating a group word library of the target group based on the word input history records of each of the objects in the target group, the input method word library generating device further comprises:

[0093] A fifth acquiring unit, configured to acquire a group name of the target group;

[0094] The second generating unit generates a group vocabulary name of the group vocabulary based on the group name and a predetermined rule.

[0095] Optionally, the fourth acquiring unit is specifically used for:

[0096] In response to triggering the group vocabulary control, displaying a detail display page of the group vocabulary, wherein the detail display page includes a link for downloading the group vocabulary;

[0097] In response to triggering the link, the group vocabulary is downloaded.

[0098] Optionally, the first display unit is specifically used for:

[0099] Obtaining a filtered group word library containing words with the same code as the first code string from the group word library set;

[0100] Determining that the target terminal of the target object contains a first application, wherein the first application meets the target terminal application benchmark of the target group corresponding to the filtered group vocabulary;

[0101] Determining that the location information of the target object matches a location information benchmark of the target group corresponding to the filtered group vocabulary;

[0102] The filtered group vocabulary is used as the group vocabulary of the target group, and a group vocabulary control of the group vocabulary is displayed.

[0103] According to one aspect of the present disclosure, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the input method vocabulary generation method as described above, or the text input method as described above when executing the computer program.

[0104] According to one aspect of the present disclosure, a computer-readable storage medium is provided, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the input method vocabulary generation method as described above, or the text input method as described above is implemented.

[0105] According to one aspect of the present disclosure, a computer program product is provided. The computer program product includes a computer program. The computer program is read and executed by a processor of a computer device, so that the computer device executes the input method word library generation method as described above.

[0106] In the disclosed embodiment, for a target group, an object feature benchmark of the target group is obtained. For each object in the input method object set, the object feature of the object is compared with the object feature benchmark of the target group to determine whether the object feature of the object matches the object feature benchmark of the target group, and the target group is formed with the matching objects. Then, based on the word input history records of each of the objects in the target group, a group vocabulary of the target group is constructed. In this way, a group vocabulary of words frequently used by objects in specific groups is established. When the word that the object wants to input in the input method has a low frequency of occurrence in the entire domain, but is a word frequently used by objects in a certain field or range, it will fall into the group vocabulary of a certain group. Combining the general vocabulary and the group vocabulary increases the probability of finding the word that the object wants to input, and improves the input efficiency of the input method.

[0107] Other features and advantages of the present disclosure will be described in the following description, and partly become apparent from the description, or understood by practicing the present disclosure. The purpose and other advantages of the present disclosure can be realized and obtained by the structures particularly pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0108] The accompanying drawings are used to provide further understanding of the technical solution of the present disclosure and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the technical solution of the present disclosure and do not constitute a limitation on the technical solution of the present disclosure.

[0109] Figure 1 is a schematic diagram of a system architecture of an embodiment of the present disclosure applied to a text input method;

[0110] Figure 2A -G is a schematic diagram of an application scenario of the text input method according to an embodiment of the present disclosure;

[0111] Figure 3 It is a flowchart of a method for generating an input method word library according to an embodiment of the present disclosure;

[0112] Figure 4 A specific flow chart for obtaining seed words is shown;

[0113] Figure 5 Shows Figure 4 Specific example diagram corresponding to the flowchart;

[0114] Figure 6 A specific flow chart for determining a first discrimination score is shown;

[0115] Figure 7 Shows Figure 6 Specific example diagram corresponding to the flowchart;

[0116] Figure 8 Shows Figure 3 A specific flow chart of step 330;

[0117] Fig. 9 Shows Figure 8 A specific flow chart of step 820 in FIG.

[0118] Fig. 10A -B shows Fig. 9 Specific example diagram corresponding to the flowchart;

[0119] Fig.11 Shows Figure 3 Another specific flow chart of step 330;

[0120] Fig.12 Shows Figure 3 Another specific flow chart of step 330;

[0121] Fig.13A -E shows Fig.12 Specific example diagram corresponding to the flowchart;

[0122] Fig.14 A specific flow chart for determining the accuracy of the first group or the second group is shown;

[0123] Fig.15A -B shows Fig.14 Specific example diagram corresponding to the flowchart;

[0124] Fig.16 Shows Figure 3 A specific flow chart of step 340;

[0125] Fig.17 Shows Fig.16 A specific flow chart of step 1610 in FIG.

[0126] Fig.18 Shows Fig.17 Specific example diagram corresponding to the flowchart;

[0127] Fig.19 Shows Fig.16 Another specific flow chart of step 1610;

[0128] Fig. 20 Shows Fig.19 A specific flow chart of step 1910 in FIG.

[0129] Fig.21 Another schematic diagram of the process of generating a word library of an input method disclosed in the present invention is shown;

[0130] Fig. 22 is a flowchart of a text input method according to an embodiment of the present disclosure;

[0131] Fig.23 Shows Fig. 22 A specific flow chart of step 2230;

[0132] Fig.24 Shows Fig. 22 A specific flow chart of step 2220;

[0133] Fig.25 A schematic diagram of the overall process of the input method word library generation method according to an embodiment of the present disclosure is shown;

[0134] Fig.26 A schematic diagram of an overall process of combining the input method word library generation method with the text input method according to an embodiment of the present disclosure is shown;

[0135] Fig. 27 is a module diagram of an input method word library generating device according to an embodiment of the present disclosure;

[0136] Fig.28 is a module diagram of a text input device according to an embodiment of the present disclosure;

[0137] Fig.29 According to the embodiment of the present disclosure Figure 3 The input method word library generation method shown, or the terminal structure diagram of the text input method;

[0138] Fig.30 According to the embodiment of the present disclosure Figure 3 The input method word library generation method shown, or the server structure diagram of the text input method. DETAILED DESCRIPTION

[0139] In order to make the purpose, technical solution and advantages of the present disclosure more clear, the present disclosure is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present disclosure and are not used to limit the present disclosure.

[0140] It should be noted that in various specific implementations of the present disclosure, when it comes to the need to perform relevant processing based on data related to the characteristics of the target object such as the target object attribute information or attribute information set, the permission or consent of the target object will be obtained first, and the collection, use and processing of these data will comply with relevant laws, regulations and standards. In addition, when the embodiment of the present disclosure needs to obtain the attribute information of the target object, the separate permission or separate consent of the target object will be obtained through a pop-up window or jump to a confirmation page. After clearly obtaining the separate permission or separate consent of the target object, the necessary target object-related data for the normal operation of the embodiment of the present disclosure will be obtained.

[0141] Before further describing the embodiments of the present disclosure in detail, the nouns and terms involved in the embodiments of the present disclosure are described. The nouns and terms involved in the embodiments of the present disclosure are subject to the following interpretations:

[0142] Object group: a group of objects that meet certain characteristics. An object group can be an object group that belongs to a certain professional field, an object group that belongs to a certain geographical location, or an object group that belongs to a certain field of interest. For example, all students in a certain university can be considered a college student group, and this group of college students can constitute an object group; for another example, the user group of a certain application can constitute an object group.

[0143] Word: The smallest unit in a language that has a fixed phonetic form and specific meaning and can be used independently, such as "game", "mobile phone", etc.

[0144] Seed words: words commonly used in a certain object group, which are different from general words. For a word class of a certain subject concept, seed words can be used as the feature items of this word class. For example, the common words "API" and "system architecture" in the programmer group are often used in the programmer group, but they are not common words in all groups, so they can be used as seed words for the programmer group; for another example, the common word "da wo" for fishing enthusiasts is often used in the fishing enthusiast group, but it is not a common word in all groups, so it can be used as a seed word for the fishing enthusiast group; for another example, for a certain university student group, the names of places on campus or near the school are often used in this part of the college student group, but they are not common words in all groups, so the names of many places in or near the university can be used as seed words for this object group.

[0145] Thesaurus: a database consisting of several words. It should be pointed out that when the user of the input method uses the input method to type, he needs to input a coded string through the input method, and then search for words in the input method's supporting thesaurus based on the coded string, and further display the searched words. Therefore, the thesaurus is the basis for searching words in the input method typing process.

[0146] Common vocabulary: refers to a vocabulary consisting of common words that are commonly used by multiple object groups.

[0147] Group vocabulary: refers to a vocabulary consisting of multiple seed words of a certain object group. One object group corresponds to one group vocabulary.

[0148] It should be noted that current input methods generally use a general word library. When generating a general word library, the frequency of input words of all objects of the input method is counted, and the number of times each input word is input by all objects of the input method is counted. The input words with the top input times are selected as the general word library. When an object uses the input method to input, the general word library is searched according to the code string (pinyin, etc.) input by the object, and the target text corresponding to the code string retrieved in the general word library is displayed for the object to select.

[0149] Today's society is developing rapidly, and people often have different text input demands. The words needed by different groups of people are often very different. Due to the capacity limitations of the general vocabulary, input methods often only store some high-frequency words. When the word entered by the object appears less frequently in the entire domain, but is a word frequently used by objects in a certain field or range, it may be difficult to find the word in the general vocabulary. In this case, the object needs to search and combine words one by one in the input method interface, resulting in low input efficiency.

[0150] System architecture and scenario description of the application of the embodiments of the present disclosure

[0151] See also Figure 1 The text input system shown, Figure 1 This is a system architecture diagram of one type of system used in the text input method according to the embodiment of the present disclosure, which includes a terminal 140, the Internet 130, a gateway 120, an input method server 110, and the like.

[0152] The terminal 140 is a device for inputting various texts, which can be input from the terminal 140 in the form of a code string. The terminal 140 includes various forms such as a desktop computer, a laptop computer, a PDA (personal digital assistant), a mobile phone, a vehicle-mounted terminal, a dedicated terminal, etc. In addition, it can be a single device or a collection of multiple devices. For example, multiple desktop computers are connected to each other through a local area network, share a display, etc. to work together, and together constitute a terminal 140. The terminal 140 can communicate with the Internet 130 in a wired or wireless manner to exchange data.

[0153] The input method server 110 refers to a computer system that can provide text input services to the terminal 140. Compared with the ordinary terminal 140, the input method server 110 has higher requirements in terms of stability, security, performance, etc. The input method server 110 can be a high-performance computer in a network platform, a cluster of multiple high-performance computers, a part of a high-performance computer (such as a virtual machine), a combination of parts of multiple high-performance computers (such as virtual machines), etc. In some embodiments, the input method server 110 includes a code string acquisition module, a word search module, a general word library, and a group word library. It should be noted that the code string acquisition module is used to obtain the code string input by the object from the terminal 140; the word search module is used to search for words in each word library based on the code string input by the object; the general word library includes the words that are more commonly used by all objects of the input method; the group word library is a word library that reflects the word usage habits of a certain object group, including the seed words that are more commonly used by a certain specific object group of the input method. It should be understood that the encoding string acquisition module, the word search module, the general word library and the group word library can be integrated in the same input method server 110, or can be deployed in different servers respectively, and are not limited to the specific embodiments listed above.

[0154] The gateway 120 is also called an internetwork connector or a protocol converter. The gateway realizes network interconnection at the transport layer and is a computer system or device that acts as a converter. The gateway is a translator between two systems that use different communication protocols, data formats or languages, or even have completely different architectures. At the same time, the gateway can also provide filtering and security functions. The message sent by the terminal 140 to the input method server 110 is sent to the corresponding input method server 110 through the gateway 120. The message sent by the input method server 110 to the terminal 140 is also sent to the corresponding terminal 140 through the gateway 120.

[0155] The embodiments of the present disclosure can be applied to a variety of scenarios. It can be applied to text input by users of an input method in a certain professional field, or to text input by users of an input method in a certain geographical location area, or to text input by users of an input method in a certain area of interest. It should be understood that the objects to which the input method thesaurus generation method and the text input method of the present disclosure can be applied are not limited to the specific embodiments listed above.

[0156] The following will refer to Figure 2A -G to illustrate some application scenarios of the present disclosure.

[0157] As Figure 2A shown, in some embodiments, the target object as the user of the input method is a member of a program enthusiasts association. Since programs are inseparable from code, members of the program enthusiasts association often habitually choose other homophones of "code" (such as: horse, mammoth) to determine some words (such as: achieving immediate success, mammoth), and further, replace the "code" in the word "code" with other homophones of "code" (such as: horse, mammoth) to create a new word (for example: achieving success in the code path, fierce code). In this way, the newly formed words can be used to name some activities in the program enthusiasts association, such as the "achieving success in the code path" plan, the "fierce code" project, and so on. Therefore, when members of the program enthusiasts association need to perform text input, they often encounter difficulties in looking up words.

[0158] As Figure 2B shown, during the process of the target object inputting a text "achieving success in the code path" in the input method interface, since this word is not a common word, if you need to type this word, you have to search and combine each character in the input method interface, change "ma'dao'cheng'gong" to "code'dao'cheng'gong", then "code path'cheng'gong", and finally to "achieving success in the code path". It should be emphasized that searching and combining each character in the input method interface will result in low text input efficiency.

[0159] As Figure 2B -C shown, during the process of the text "achieving success in the code path" being input to the terminal display interface, using "ma'dao'cheng'gong" as the first coding string corresponding to the target text, and using the input method to determine each character to be typed one by one: from "code'dao'cheng'gong" to "code path'cheng'gong" and then to "achieving success in the code path". When the input method detects that "ma'dao'cheng'gong" is input by the target object, or detects that "achieving success in the code path" is typed out character by character by the target object, it will push the group word library control "program enthusiasts association thesaurus" corresponding to "achieving success in the code path" to the target object through the input method.

[0160] As Figure 2D shown, when the grouped word library control is triggered by the target object, the details page will immediately pop up on the input method interface. The details page presents the word library recommended based on "Code Path Success", that is, the "Program Enthusiasts Association Word Library", which includes vocabulary related to programs and vocabulary customized by the Program Enthusiasts Association. Among them, the vocabulary related to programs are: application program; application, application program, application program framework, application framework, application program framework, architecture, system architecture, system structure, array, array, arrow operator, assembly language, assembly language, setting, assignment, associative container, atomicity, artificial intelligence, artificial intelligence, code, etc.; the vocabulary customized by the Program Enthusiasts Association includes: Code Path Success, Code Path Research Group, Code Path Planning, Mighty Code, etc.

[0161] As Figure 2E shown, according to some other embodiments provided by the present disclosure, during the process of inputting the text "go to X Dynasty" into the terminal display interface, "fu'X'zhao" is used as the first coding string corresponding to the target text, and the input method is used to determine the characters to be typed one by one: from "go to 'X' zhao" to "go to X Dynasty". When the input method detects that "fu'X'zhao" is input by the target object, or detects that "go to X Dynasty" is typed out character by character by the target object, the grouped word library control "A Game Word Library" corresponding to "go to X Dynasty" will be pushed to the target object through the input method.

[0162] As Figure 2F shown, when the grouped word library control is triggered by the target object, the details page will immediately pop up on the input method interface. The details page presents the word library recommended based on "go to X Dynasty", that is, the "A Game Association Word Library", which includes common game terms and vocabulary customized by game producers. Among them, the common game terms are: mid lane, necklace, item, equipment, canyon, river, skill, skill point, attribute, attack speed, cooldown time, movement speed, ultimate skill, ultimate move, heal, flash, teleport, talent, etc.; the vocabulary customized by game producers includes: go to X Dynasty, Han X, Hou X, X Bai, Da X, Zhong X, Lv X, etc.

[0163] As Figure 2GAs shown, in addition to selecting the corresponding group vocabulary by entering the first encoding string corresponding to the target text, or typing the target text so that the input method pushes a group vocabulary control to it, the target object can also directly search for the corresponding group vocabulary in the alternative vocabulary pool that comes with the input method. It should be noted that the alternative vocabulary pool includes a variety of alternative group vocabulary. After the target object enters the keyword corresponding to the target text in the search bar, the search button can be clicked to search for the group vocabulary corresponding to the target text. It should be noted that when the target object does not use keyword search, it can also select a vocabulary that meets the needs of the target object from the classification of different group vocabulary. The classification of group vocabulary can include: mobile games, animation and comics, sports, economics and finance, medicine and pharmaceuticals, construction engineering, literature and art, study exams, and the Internet, among other categories. In addition, the input method can also recommend some group vocabulary to the target object for the target object to choose from, for example Figure 2G The "A game vocabulary" and "game field vocabulary" shown on the recommendation page.

[0164] It should be clear that the group word library involved above is generated by the input method word library generation method provided by the embodiment of the present disclosure. It should be understood that the above content only illustrates the description of some application scenarios of the present disclosure, and the business scenarios to which the present disclosure can be applied may include, but not be limited to, the specific embodiments listed above.

[0165] General description of the disclosed embodiments

[0166] It should be emphasized that the embodiments of the present disclosure can be applied to a variety of scenarios, and can be applied to input method users in a certain professional field for text input, can be applied to input method users in a certain geographical area for text input, and can be applied to input method users in a certain field of interest for text input. It should be understood that the objects to which the input method word library generation method and text input method disclosed in the present disclosure can be applied are not limited to the specific embodiments listed above.

[0167] According to some embodiments of the present disclosure, an input method word library generation method, a text input method, related devices and media are provided, which can improve the text input efficiency of the input method.

[0168] Detailed description of step 300

[0169] Reference Figure 3 The input method word library generation method of an embodiment of the present disclosure may include, but is not limited to, the following steps 310 to 340.

[0170] Step 310, obtaining object feature benchmarks of the target group;

[0171] Step 320, obtaining the object feature of each object in the object set using the input method;

[0172] Step 330, in the object set, objects whose object features match the object feature benchmark are obtained to form a target group;

[0173] Step 340 , based on the word input history records of each object in the target group, generate a group word library of the target group for use in the input method.

[0174] Steps 310 to 340 are described in detail below.

[0175] In step 310, the object feature benchmark of the target group is obtained. It should be emphasized that the object group is a group of objects that meet certain characteristics, the seed word is a common word in a certain object group that is different from the general vocabulary, and the group vocabulary refers to a vocabulary composed of multiple seed words of a certain object group, and one object group corresponds to one group vocabulary. It should be noted that the target group is an object group used to provide a target for the establishment of a group vocabulary.

[0176] In some exemplary embodiments of the present disclosure, in order to construct a group vocabulary, it is necessary to first determine which objects are specifically included in the target group. It should be noted that the object feature benchmark is used to define the common features of multiple objects in the target group. Therefore, the purpose of obtaining the object feature benchmark of the target group is to facilitate finding objects with common object features from the object set using the input method based on the object feature benchmark in subsequent steps, so as to facilitate determining the target group.

[0177] In some specific embodiments of the present disclosure, the object feature benchmark can be obtained through Internet transmission, through an input device configured in a terminal, or through other methods. It should be understood that there are many ways to obtain the object feature benchmark, not limited to the above examples.

[0178] In step 320, the object features of each object in the object set using the input method are obtained. It should be noted that the object features are used to characterize a certain characteristic of an object. It should be noted that the object set of the input method refers to a collection of multiple objects that have been input text using the input method. It should be pointed out that each object in an object group needs to have a common object feature. Therefore, in order to facilitate comparison with the object feature benchmark, it is necessary to obtain the object features of each object in the object set using the input method.

[0179] In some more specific embodiments, the types of object features are various, and may be an application installed on the object terminal of the object, a word input history record of the object, or a location information of the object.

[0180] In some embodiments, when the object feature is an application installed on the object terminal of the object, the object feature can be obtained in the following manner: by means of a pop-up window or jumping to a confirmation page, requesting permission to "query the local application list" to obtain the object's separate permission or consent. After the separate permission or consent of the target object is clearly obtained, that is, the permission to "query the local application list" is obtained, the application installed on the object terminal can be queried from the local application list.

[0181] In some embodiments, when the object feature is the word input history of the object, the object feature can be obtained in the following manner: by means of a pop-up window or jumping to a confirmation page, requesting permission to obtain the "query word input history" to obtain the separate permission or consent of the object. After the separate permission or consent of the target object is clearly obtained, that is, the permission of the "query word input history" is obtained, the word input history of the object can be queried and obtained.

[0182] In some embodiments, when the object feature is the location information of the object, the object feature can be obtained in the following manner: by means of a pop-up window or jumping to a confirmation page, requesting permission to obtain "query location information" to obtain separate permission or consent from the object. After the separate permission or consent of the target object is clearly obtained, that is, after the permission to "query location information" is obtained, the location information of the object can be queried and obtained.

[0183] It should be understood that there are various ways to obtain object features and are not limited to the specific embodiments listed above.

[0184] In step 330, objects whose object features match the object feature benchmark are obtained from the object set to form a target group. It should be noted that when the object features of an object in the object set are compared with the object feature benchmark and it is determined that the object features match the object feature benchmark, the object can be used as a member of the target group. Therefore, multiple objects whose object features match the object feature benchmark are obtained from the object set, and a target group can be formed based on the multiple objects.

[0185] In step 340, based on the word input history records of each object in the target group, a group vocabulary of the target group is generated for the input method. It should be noted that after the target group is formed, the word input history records of each object in the target group can be further extracted, and then the word input history records of each object in the target group can be integrated to generate the group vocabulary of the target group. It should be clear that after the group vocabulary of the target group is generated, when the objects belonging to the target group use the input method to input text, the group vocabulary of the target group can be used to look up words. In this way, even if the words input by the object have a low frequency of occurrence in the whole domain, the words frequently used by objects in a certain field or range no longer need to be searched and combined word by word in the input method interface, thereby improving the efficiency of text input.

[0186] According to the disclosed embodiment shown in step 310 to step 340, for a target group, an object feature benchmark of the target group is obtained. For each object in the input method object set, the object feature of the object is compared with the object feature benchmark of the target group to determine whether the object feature of the object matches the object feature benchmark of the target group, and the target group is formed with the matching objects. Then, based on the word input history records of each of the objects in the target group, a group vocabulary of the target group is constructed. In this way, a group vocabulary of words frequently used by objects in specific groups is established. When the word that an object wants to input in the input method has a low frequency of occurrence in the entire domain, but is a word frequently used by objects in a certain field or range, it will fall into the group vocabulary of a certain group. Combining the general vocabulary and the group vocabulary increases the probability of finding the word that the object wants to input, and improves the input efficiency of the input method.

[0187] Reference Figure 4 It should be emphasized that seed words are common words in a certain object group that are different from common words in a certain object group. In some embodiments, seed words can be obtained by following steps 410 to 460:

[0188] Step 410, obtaining a first subgroup and a second subgroup, wherein the first subgroup includes a plurality of first objects that are predetermined to belong to a target group, and the second subgroup includes a plurality of second objects that are predetermined to not belong to the target group;

[0189] Step 420, obtaining a first corpus, the first corpus including a plurality of first sample corpora of first objects;

[0190] Step 430, obtaining a second corpus, where the second corpus includes a plurality of second sample corpora of second objects;

[0191] Step 440, segment each first sample corpus in the first corpus set to obtain candidate words;

[0192] Step 450, for each candidate word, determining the first number of times the candidate word appears in the first corpus and the second number of times the candidate word appears in the second corpus;

[0193] Step 460, determining a seed word from the candidate words based on the first number and the second number.

[0194] Steps 410 to 460 are described in detail below.

[0195] In step 410, a first seed group and a second seed group are obtained, the first seed group includes a plurality of first objects that are predetermined to belong to the target group, and the second seed group includes a plurality of second objects that are predetermined not to belong to the target group. It should be noted that in order to determine the common words that are different from the common vocabulary in the target group, it is necessary to determine in advance the plurality of first objects that belong to the target group, and integrate the plurality of first objects that belong to the target group into the first seed group; in addition, it is also necessary to determine in advance the plurality of second objects that do not belong to the target group, and integrate the plurality of second objects that do not belong to the target group into the second seed group. It should be pointed out that the purpose of obtaining the first seed group and the second seed group is to facilitate the determination of seed words by parsing the corpus of the first seed group and the corpus of the second seed group in subsequent steps.

[0196] In step 420 to step 430, a first corpus is obtained, the first corpus includes a plurality of first sample corpora of the first object; and a second corpus is obtained, the second corpus includes a plurality of second sample corpora of the second object. It should be noted that the first corpus is a collection of corpora input by the first object through the input method, so the first corpus includes a plurality of first sample corpora of the first object; similarly, the second corpus is a collection of corpora input by the second object through the input method, so the second corpus includes a plurality of second sample corpora of the second object.

[0197] In some embodiments, the text input by multiple objects using the input method of the embodiment of the present disclosure will be attached with the identification information of the object. In this way, the corpus of the object can be sorted out from the text corpus of the input method based on the identification information of a certain object. Therefore, in some embodiments, after obtaining the first seed group and the second seed group, the first corpus can be obtained from the identification information based on multiple first objects; similarly, the second corpus can also be obtained from the identification information based on multiple second objects.

[0198] In some more specific embodiments, when an object enters text after logging into an account on an input method, the object's account information will be attached to the input text. In this embodiment, the account information is identification information.

[0199] In step 440, each first sample corpus in the first corpus set is segmented to obtain candidate words. It should be noted that word segmentation is the basis of natural language processing, and the accuracy of word segmentation directly determines the quality of candidate words. It should be pointed out that candidate words are words that are candidates to become seed words, and seed words are selected from multiple candidate words.

[0200] In some more specific embodiments, since English sentences use spaces to separate words, except for some specific words (such as how many, New York, etc.), word segmentation does not need to be considered in most cases. However, Chinese is different. It naturally lacks separators and requires readers to segment words and sentences by themselves. Therefore, when doing Chinese natural language processing, we need to perform word segmentation first.

[0201] In some other more specific embodiments, for Chinese word segmentation, the current word segmentation methods are mainly divided into two categories, the dictionary-based rule matching method, and the statistical-based machine learning method. First, the dictionary-based word segmentation algorithm is essentially string matching. The string to be matched is matched with a sufficiently large dictionary based on a certain algorithm strategy. If the match is successful, the word segmentation can be performed. According to different matching strategies, it is divided into forward maximum matching method, reverse maximum matching method, two-way matching word segmentation, full segmentation path selection, etc.; second, the statistical-based word segmentation algorithm is essentially a sequence labeling problem. In some embodiments, the words in the sentence can be labeled according to their position in the word. The main labels are: B (a word at the beginning of the word), E (the last word of the word), M (a word in the middle of the word, possibly multiple), S (a word represented by a word). For example, "Today's weather is really good", the result after labeling is "BESBESBEBME", and the corresponding word segmentation result is "Today / the / weather / is / really / very good".

[0202] It should be understood that there are various methods for word segmentation, which may be word-by-word segmentation, segmentation according to Chinese vocabulary, or other segmentation methods. Therefore, the word segmentation process for the first sample corpus is not limited to the specific embodiments listed above.

[0203] In step 450 to step 460, for each candidate word, the first number of times the candidate word appears in the first corpus and the second number of times the candidate word appears in the second corpus are determined, and then based on the first number and the second number, the seed word is determined from the candidate word. It should be emphasized that the seed word is a common word of an object group in a certain object group that is different from the general vocabulary, and the first object belongs to the target group, and the second object is deployed in the target group. Therefore, the seed word of the target group is more common in the first corpus, that is, the first number is larger, and it is less common in the second corpus, that is, the second number is smaller.

[0204] Reference Figure 5 , the present disclosure provides some embodiments. Among them, the first subgroup includes a plurality of first objects such as a first object a, a first object b, a first object c, and a first object d. The text corpus input by the first object a using the input method is the first sample corpus a, the text corpus input by the first object b using the input method is the first sample corpus b, the text corpus input by the first object c using the input method is the first sample corpus c, and the text corpus input by the first object d using the input method is the first sample corpus d; similarly, the second subgroup includes a plurality of second objects such as a second object a, a second object b, a second object c, and a second object d. The text corpus input by the second object a using the input method is the second sample corpus a, the text corpus input by the second object b using the input method is the second sample corpus b, the text corpus input by the second object c using the input method is the second sample corpus c, and the text corpus input by the second object d using the input method is the second sample corpus d. Furthermore, the first corpus set including multiple first sample corpora such as first sample corpus a, first sample corpus b, first sample corpus c, first sample corpus d, etc. is segmented to obtain multiple candidate words such as [candidate word a1], [candidate word a2], [candidate word b1], [candidate word b2].

[0205] Furthermore, for each candidate word, determine the first number of times the candidate word appears in the first corpus, and the second number of times the candidate word appears in the second corpus. Among them, the first number a1 of [candidate word a1] in the first corpus, the second number a1 of [candidate word a1] in the second corpus, the first number a2 of [candidate word a2] in the first corpus, and the second number a2 of [candidate word a2] in the second corpus; the first number b1 of [candidate word b1] in the first corpus, the second number b1 of [candidate word b1] in the second corpus, the first number b2 of [candidate word b2] in the first corpus, and the second number b2 of [candidate word b2] in the second corpus. It should be understood that the first and second numbers corresponding to other candidate words are in Figure 5 The expression form in is consistent with the expression forms of [candidate word a1], [candidate word a2], [candidate word b1], and [candidate word b2] mentioned above, and will not be repeated here.

[0206] The present disclosure provides some embodiments, which require determining the first number of times a candidate word appears in a first corpus and the second number of times the candidate word appears in a second corpus. Further, based on the first number and the second number, a seed word is determined from the candidate word.

[0207] In some embodiments, to determine a seed word from candidate words based on the first count and the second count, the first count can be divided by the second count, and then, according to the magnitude of the quotient, the seed word can be determined from the candidate words. For example, when the first count of the candidate word "code path success" appearing in the first corpus is 80 and the second count of it appearing in the second corpus is 10, the result of the division is 8; when the first count of the candidate word "achieve immediate victory" appearing in the first corpus is 20 and the second count of it appearing in the second corpus is also 20, the result of the division is 1; similarly, the division of the first count by the second count is performed for each of the other candidate words one by one. In this way, the candidate word with a higher quotient result is more suitable as a seed word. It should be noted that for multiple candidate words, the quotient results are calculated in the above manner, and then the candidate words with quotient results higher than a certain predetermined threshold can be determined as seed words.

[0208] It should be noted that based on the first count and the second count corresponding to a candidate word, the seed word can also be determined without relying on the quotient result of the two. For example, the candidate words are directly arranged in descending order according to the first count to obtain a first candidate word sequence, and then the first N candidate words are selected from the first candidate word sequence. Further, these N candidate words are rearranged in descending order according to the second count to obtain a second candidate word sequence, and the first M candidate words are selected from the second candidate word sequence, and these M candidate words are determined as seed words.

[0209] In some other embodiments, the first word frequency corresponding to the first count can be determined first based on the ratio of the first count to the total number of all words in the first corpus; similarly, the second word frequency corresponding to the second count can be determined based on the ratio of the second count to the total number of all words in the second corpus. Then, a batch of candidate words with a larger first word frequency and a smaller second word frequency are determined, and further, this batch of candidate words are determined as seed words.

[0210] It should be understood that the seed words can be obtained in a variety of ways, which can include, but are not limited to, the specific embodiments cited above.

[0211] It should be clear that through the embodiments of the present disclosure shown in steps 410 to 460, higher-quality seed words can be obtained, so as to facilitate the acquisition of the object feature benchmark, thereby generating a more accurate group word library for the target group, and further improving the input efficiency corresponding to the input method.

[0212] Refer to Figure 6 , the first discrimination score of the seed word is used to quantify the discrimination effect of the seed word. In some embodiments, the first discrimination score can be obtained through the following steps 610 to 640.

[0213] Step 610, obtain a third corpus, where the third corpus includes sample corpora of the first number of third objects;

[0214] Step 620, obtaining the third occurrence number of the seed word in the third corpus;

[0215] Step 630, obtaining a second number, wherein in the third corpus, sample corpora of a second number of objects contain the seed word, and the second number is not greater than the first number;

[0216] Step 640, determining a first discrimination score based on the third number and the second number.

[0217] Steps 610 to 640 are described in detail below.

[0218] It should be emphasized that the seed word is a common word in a certain object group that is different from the common words in the object group. Therefore, in order to measure the distinguishing effect of the seed word, it is necessary to calculate the first distinguishing score for the seed word through steps 610 to 640.

[0219] In step 610, a third corpus is obtained, and the third corpus includes sample corpora of the first number of third objects. It should be noted that the third corpus is a corpus that includes seed words. The third corpus includes sample corpora of the first number of third objects.

[0220] In step 620 to step 630, the third number of occurrences of the seed word in the third corpus is obtained; the second number is obtained, wherein in the third corpus, the sample corpus of the second number of objects contains the seed word, and the second number is not greater than the first number. It should be noted that the seed word has the following characteristics: the seed word does not appear frequently in the third corpus, but it appears frequently in the sample corpus of some objects; it should be pointed out that the seed word is often used by some objects but not widely used by all objects, which is a manifestation of a high distinguishing effect. In order to quantify the distinguishing effect of the seed word based on this method, it is necessary to obtain the third number of occurrences of the seed word in the third corpus, and obtain the number of objects whose sample corpus contains the seed word, that is, the second number. In this way, the third number and the second number can provide a data basis for determining the first discrimination score.

[0221] In step 640, a first discrimination score is determined based on the third number and the second number. It should be emphasized that the third number is the frequency of the seed word appearing in the third corpus, and the second number is the number of objects in the sample corpus that contain the seed word. Therefore, determining the first discrimination score based on the third number and the second number can be achieved in a variety of ways.

[0222] In some more specific embodiments, determining the first discrimination score based on the third number and the second number may include:

[0223] Get the first constant;

[0224] A first discrimination score is determined based on the logarithm of the third number, the first constant, and the logarithm of the second number.

[0225] It should be noted that in some embodiments, the seed word is represented as w, the first constant is 100, the third number is represented as tf(w), and the second number is represented as uf(w). On this basis, the first discrimination score Score_tfidf(w) can be expressed as:

[0226] Score_tfidf(w)=log(tf(w))*100 / log(uf(w))

[0227] It should be clear that the first discrimination score Score_tfidf(w) calculated in the above manner can quantify the discriminative effect of the seed word. The higher the first discrimination score Score_tfidf(w), the more it can be said that the seed word is often used by some objects but not widely used by all objects. Therefore, it helps to accurately measure the discriminative effect of the seed word.

[0228] It should be understood that based on the third digit and all the words in the third corpus, the word frequency of the third digit can be calculated, and then the first discrimination score is determined according to the word frequency of the third digit and the second digit. Therefore, in some embodiments, tf(w) in the expression of the first discrimination score Score_tfidf(w) can be either the third digit or the word frequency corresponding to the third digit.

[0229] Reference Figure 7, in some other more specific embodiments, a method for calculating the first discrimination score is also shown. First, it is necessary to obtain the third corpus set, which contains 6 sample corpora of the third object, namely: sample corpus a of the third object, sample corpus b of the third object, sample corpus c of the third object, sample corpus d of the third object, sample corpus e of the third object and sample corpus f of the third object. It should be clear that the seed word appears 30 times in the sample corpus a of the third object, 25 times in the sample corpus b of the third object, 0 times in the sample corpus c of the third object, 40 times in the sample corpus d of the third object, 0 times in the sample corpus e of the third object, and 45 times in the sample corpus f of the third object. Therefore, the third number of times the seed word appears in the third corpus is: 30+25+0+40+0+45=140 times. Since among the above 6 sample corpora of the third object, only the sample corpora a of the third object, the sample corpora b of the third object, the sample corpora d of the third object and the sample corpora f of the third object contain seed words, the second number is 4. In this way, the third number is determined to be 140 and the second number is 4. Further, the first discrimination score can be directly determined according to the ratio of the third number to the second number: 140 / 4=35.

[0230] In some other embodiments, the first discrimination score may also be calculated based on the word frequency corresponding to the third number in the third corpus. For example, if the third corpus contains a total of 1000 words, the first discrimination score Score_tfidf(w) is calculated based on the third number 140 corresponding to the seed word, the second number 4, and the constant 100, expressed as:

[0231] Score_tfidf(w)=log(140)*100 / log(4)=356.5

[0232] It should be understood that there are various ways to determine the first discrimination score, which may include, but not be limited to, the specific embodiments listed above.

[0233] It should be clear that through the embodiments of the present disclosure shown in 610 to 640, the distinguishing effect of the seed word can be quantified more reasonably through the first discrimination score, which helps to obtain higher quality seed words, facilitates the acquisition of object feature benchmarks, and thus generates a more accurate group vocabulary for the target group, further improving the input efficiency corresponding to the input method.

[0234] Detailed description of step 330

[0235] Reference Figure 8The object feature benchmark includes the object terminal application benchmark, the seed word benchmark and the positioning information benchmark; the object feature includes the application installed on the object terminal, the object's word input history, and the object's positioning information. Step 330 may include, but is not limited to, the following steps 810 to 840.

[0236] Step 810, for an object in the object set, determining that an application installed on an object terminal of the object meets an object terminal application benchmark;

[0237] Step 820, determining that the word input history record of the object meets the seed word benchmark;

[0238] Step 830, determining whether the positioning information of the object matches the positioning information reference;

[0239] Step 840, adding the object to the target group.

[0240] Steps 810 to 840 are described in detail below.

[0241] It should be noted that obtaining objects whose object features match the object feature benchmark to form a target group can specifically include three types of comparison processes: first, the comparison process of the application; second, the comparison process of the word input history record; and third, the comparison process of the positioning information.

[0242] In step 810, for an object in the object set, it is determined whether the application installed on the object terminal of the object meets the object terminal application benchmark. It should be noted that the terminal usage characteristics of the object can be determined through the application on the object terminal. If the object terminal has an application that matches the object terminal application benchmark installed, it means that the owner of the object terminal has the common characteristics of the target group in terminal usage.

[0243] In some more specific embodiments, the objects corresponding to the college student group of a certain school often install some applications for handling school affairs on their object terminals. These applications can be used to handle matters such as course attendance and facility repair. Therefore, if such applications are installed on the object terminal of a certain object, it can play a certain role in defining its identity characteristics. In this way, it means that the owner of the object terminal can be added to the target group corresponding to the college student group of a certain school.

[0244] Step 820, determine whether the subject's word input history record matches the seed word benchmark. It should be noted that by comparing the word input history record of the input method, the subject's text input habits can be determined. If the word input history record of a subject matches the seed word benchmark, it means that the subject has common characteristics in the text input habits of the target group.

[0245] In some more specific embodiments, certain objects who are members of the Program Enthusiast Association, since programs are inseparable from code, often use "code" to replace other homophonic words, thus creating some seed words, such as "code path to success", "fierce code" and the like. Therefore, if an object types out these seed words like "code path to success" or "fierce code" using an input method, that is, the word input history record of the object conforms to the seed word benchmark, it can play a certain role in defining its identity characteristics. In this way, it means that the object that types out these seed words can be added to the target group corresponding to the members of the Program Enthusiast Association.

[0246] In some other more specific embodiments, the development teams of some games often name these themed activities in order to hold some interesting themed activities in the games. For example, the development team of Game A names the ongoing themed activity as "going to the X dynasty". Therefore, if an object types out the seed word "going to the X dynasty" using an input method, that is, the word input history record of the object conforms to the seed word benchmark, it can play a certain role in defining its identity characteristics. In this way, it means that the object that types out these seed words can be added to the target group corresponding to the players of Game A.

[0247] In step 830, it is determined that the positioning information of the object matches the positioning information benchmark. It should be noted that by comparing the positioning information, the usual activity area of the object can be determined. If the positioning information of an object matches the positioning information benchmark, it means that the object has a common characteristic with the target group in the usual activity area.

[0248] In some more specific embodiments, permissions for "querying positioning information" are requested by means of pop-up windows or jumping to a confirmation page, etc., to obtain the separate permission or separate consent of the object. After clearly obtaining the separate permission or separate consent of the target object, that is, after the permission for "querying positioning information" is obtained, the positioning information of the target object can be queried and obtained. Further, the positioning information of the object can be compared with the positioning information benchmark. Among them, the positioning information benchmark can be a pre-defined location area, such as the area of a certain school or the area of a certain shopping mall. If the positioning information of the target object is within the pre-defined location area, it is determined that the positioning information of the object matches the positioning information benchmark; in some other embodiments, the positioning information benchmark can also be the location information of other objects in the corresponding area. The distance between the positioning information of the target object and the location information of other objects already in the corresponding area is measured, and then if the distance is within the predetermined threshold, it is determined that the positioning information of the object matches the positioning information benchmark.

[0249] Step 840: adding the object to the target group. It should be noted that after the above three types of comparison processes, objects whose object features match the object feature benchmark can be determined, thereby forming a target group.

[0250] Through the embodiment of the present disclosure shown in steps 810 to 840, based on the three types of comparison processes of the above terminal usage characteristics, text input habits and normal activity areas, it can be more comprehensively determined whether the object feature benchmark matches the object features of the target group, thereby improving the grouping accuracy of the target group, generating a more accurate group vocabulary for the target group, and further improving the input efficiency corresponding to the input method.

[0251] It should be understood that some embodiments provided by the present disclosure may include the above three types of comparison processes, may include any one of the three types of comparison processes, and may also include a process of selecting any two types from the three types for comparison.

[0252] Reference Fig. 9 The seed word benchmark includes the seed word and the first discrimination score corresponding to the seed word. Step 820 may include, but is not limited to, the following steps 910 to 930.

[0253] Step 910, setting a counter, the counter is initialized to 0;

[0254] Step 920, traversing the word input history records of the object, if a word input history record corresponds to a seed word, accumulating the first discrimination score corresponding to the seed word into a counter;

[0255] Step 930, after traversing the word input history of the object, based on the value of the counter, determine that the word input history of the object meets the seed word benchmark.

[0256] Steps 910 to 930 are described in detail below.

[0257] In step 910, a counter is set and initialized to 0. It should be noted that the counter in the present disclosure may refer to a counter module. It should be understood that various types of object terminals have a counting function. It should be noted that the purpose of setting the counter is to accumulate the first discrimination score corresponding to the seed word to determine whether the word input history record of the object matches the seed word benchmark.

[0258] In step 920, the word input history of the object is traversed. If a word input history record corresponds to a seed word, the first discrimination score corresponding to the seed word is accumulated to the counter. It should be noted that the word input history record of the object includes a variety of words. If a word input history record determines a seed word, it means that this seed word plays a role in distinguishing the object of the word input history record. It should be emphasized that the first discrimination score of the seed word is used to quantify the distinguishing effect of the seed word. The specific method of obtaining the first discrimination score can refer to the above Figure 6 A series of embodiments represented by corresponding steps. It should be noted that when referring to the above Figure 6 When the first discrimination score is obtained by a series of embodiments represented by corresponding steps, the corresponding third corpus can be extracted from the entire word input history record of the object using the input method to input text, or can be extracted from a partial word input history record of the object using the input method to input text. It should be noted that text input habits may change over time, so in order to improve the efficiency of obtaining the first discrimination score, the third corpus can also be extracted from the text input records of a certain period of time in the input method.

[0259] In step 930, after traversing the word input history of the object, based on the value of the counter, it is determined whether the word input history of the object meets the seed word benchmark. It should be noted that after traversing the word input history of the object, the sum of the first discrimination scores of all seed words in the word input history will be recorded in the counter. Therefore, it can be further determined whether the word input history of the object meets the seed word benchmark based on the sum of the first discrimination scores of all seed words in the word input history.

[0260] like Fig. 10AAs shown, in some more specific embodiments, the seed word is specifically [Code Path Success]. The word input history records of object A respectively include the sample corpus a1, sample corpus a2, sample corpus a3, sample corpus a4, sample corpus a5, and sample corpus a6 of object A. To determine whether the word input history record of the object conforms to the seed word benchmark, it is necessary to first set a counter, and the counter is initialized to 0. Then traverse the word input history record of the object. If a word input history record corresponds to a seed word [Code Path Success], add the first discrimination score corresponding to the seed word [Code Path Success] to the counter, and the total sum of the first discrimination scores of the seed word [Code Path Success] can be obtained: 30 + 10 + 25 + 25 = 90. When the word input history record of the object is traversed, based on the value of the counter, it is determined that the word input history record of the object conforms to the seed word benchmark. Among them, if the seed word benchmark is "the total sum of the first discrimination scores reaches 60", then compare the total sum of the first discrimination scores of the seed word [Code Path Success], which is 90, with "the total sum of the first discrimination scores reaches 60", and it can be determined that the word input history record conforms to the seed word benchmark.

[0261] As Fig. 10B shown, in some more specific embodiments, the seed word is specifically [Go to X Dynasty]. The word input history records of the input method respectively include the sample corpus b1, sample corpus b2, sample corpus b3, sample corpus b4, sample corpus b5, and sample corpus b6 of object B. To determine whether the word input history record of the object conforms to the seed word benchmark, it is necessary to first set a counter, and the counter is initialized to 0. Then traverse the word input history record of the object. If a word input history record corresponds to a seed word [Go to X Dynasty], add the first discrimination score corresponding to the seed word [Go to X Dynasty] to the counter, and the total sum of the first discrimination scores of the seed word [Go to X Dynasty] can be obtained: 30 + 25 + 25 = 80. When the word input history record of the object is traversed, based on the value of the counter, it is determined that the word input history record of the object conforms to the seed word benchmark. Among them, if the seed word benchmark is "the total sum of the first discrimination scores reaches 60", then compare the total sum of the first discrimination scores of the seed word [Go to X Dynasty], which is 80, with "the total sum of the first discrimination scores reaches 60", and it can be determined that the word input history record conforms to the seed word benchmark.

[0262] In some more specific embodiments, the word input history of a certain object may include multiple seed words. For multiple seed words, a corresponding first discrimination score can be obtained for each of them. Then, all the first discrimination scores are added up and compared with a pre-determined seed word benchmark to determine whether the word input history conforms to the seed word benchmark. For example, it has been determined that the seed words corresponding to the target group of college students in a certain university are [in front of Purple X], [X Lotus Poetry Society], [Qing X Garden]; calculate the first discrimination scores of each seed word in the word input history of this object as follows: [in front of Purple X], 128; [X Lotus Poetry Society], 300; [Qing X Garden], 200. In this way, the first discrimination scores of each seed word can be further summed up: 128 + 300 + 200 = 628. Further, compare the sum 628 of the first discrimination scores of each seed word with the pre-determined threshold 600, and it can be clearly seen that the sum 628 of the first discrimination scores of each seed word is greater than the pre-determined threshold 600. Therefore, it can be finally determined that the word input history conforms to the seed word benchmark.

[0263] It should be understood that there are various ways to determine whether the word input history of an object conforms to the seed word benchmark, which may include, but are not limited to, the specific embodiments cited above.

[0264] Through the embodiments of the present disclosure shown in steps 910 to 930, by accumulating and counting the first discrimination scores, it is possible to more accurately determine whether the word input history of an object conforms to the seed word benchmark, thereby further improving the grouping accuracy of the target group, generating a more accurate group vocabulary for the target group, and further improving the input efficiency corresponding to the input method.

[0265] Refer to Fig.11 , the object feature benchmark includes an object terminal application benchmark, a seed word benchmark, and a positioning information benchmark; object features include the applications installed on the object's object terminal, the word input history of the object, and the positioning information of the object. Step 330 may include, but is not limited to, the following steps 1110 to 1150.

[0266] Step 1110, for the objects in the object set, determine the first matching degree between the applications installed on the object's object terminal and the object terminal application benchmark;

[0267] Step 1120, determine the second matching degree between the word input history of the object and the seed word benchmark;

[0268] Step 1130, determine the third matching degree between the positioning information of the object and the positioning information benchmark;

[0269] Step 1140, based on the first matching degree, the second matching degree, and the third matching degree, determine the matching degree between the object feature and the object feature benchmark;

[0270] Step 1150 , based on the matching degree, obtaining an object whose object features match the object feature benchmark.

[0271] Steps 1110 to 1150 are described in detail below.

[0272] It should be emphasized that in the object set, objects whose object features match the object feature benchmark are obtained to form a target group. Specifically, three types of comparison processes can be included: first, the comparison process of the application; second, the comparison process of the word input history record; third, the comparison process of the positioning information.

[0273] In step 1110, for an object in the object set, a first degree of matching between an application installed on an object terminal of the object and an application benchmark of the object terminal is determined. It should be noted that the first degree of matching is used to measure the degree of matching between an application installed on an object terminal of the object and an application benchmark of the object terminal, and is obtained during the comparison process of the application programs. It should be noted that there are various ways to determine the first degree of matching.

[0274] In some more specific embodiments, you can first request permission to "query the local application list" through a pop-up window or jump to a confirmation page, so as to obtain the individual permission or consent of the object. After the individual permission or consent of the target object is clearly obtained, that is, after the permission to "query the local application list" is obtained, the applications installed on the object terminal can be queried based on the local application list. Among them, each time an application that matches the object terminal application benchmark is queried, the corresponding score is accumulated; after traversing the local application list, the accumulated score can be determined as the first matching degree based on the object terminal application benchmark.

[0275] In some more specific embodiments, for a group of college students in a certain school, the application used to handle school affairs has a greater distinguishing effect. When the application used to handle school affairs is queried in the local application list, it means that the object is likely to belong to the group of college students in this school, so it can get a higher score; the application used to provide online course videos may be the same video software in multiple schools, which can only effectively divide the object group into the group of college students, but it cannot effectively distinguish which specific school it is. Therefore, the application that provides online course videos has a general distinguishing effect, and the corresponding score configured for it will be lower than the application used to handle school affairs. Similarly, according to the distinguishing effect of each application from large to small, it is scored in the object terminal application benchmark, and the corresponding total score can be obtained after traversing the local application list. In this way, the total score corresponding to the local application list can be determined as the first matching degree.

[0276] In other more specific embodiments, for a group of employees of a certain company, they often use "A communication software", "B document processing software" and "C attendance software" together for office work. When one of "A communication software", "B document processing software" and "C attendance software" is found in the local application list, the corresponding score of the application can be marked for it. When the combination of any two of "A communication software", "B document processing software" and "C attendance software" is found in the local application list, the combined score is added to the sum of the corresponding scores of the two applications. When the combination of three of "A communication software", "B document processing software" and "C attendance software" is found in the local application list, a higher combined score is added to the sum of the corresponding scores of the three applications. In this way, the sum of the corresponding scores of the application programs and the combined score can be determined as the first matching degree.

[0277] It should be emphasized that there are various ways to determine the first matching degree, which may include but are not limited to the specific embodiments listed above.

[0278] In step 1120, a second degree of matching between the word input history record of the object and the seed word benchmark is determined. It should be noted that the second degree of matching is used to measure the degree of matching between the word input history record and the seed word benchmark, and is obtained during the comparison process of the word input history record. It should be noted that there are various ways to determine the second degree of matching.

[0279] In some more specific embodiments, you can first request permission to obtain the "query word input history" through a pop-up window or jump to a confirmation page, so as to obtain the individual permission or consent of the object. After the individual permission or consent of the target object is clearly obtained, that is, after the permission of the "query word input history" is obtained, you can query and obtain the word input history of the object. Among them, each time a seed word that matches the seed word benchmark is queried, the corresponding score is accumulated; after traversing the word input history, the accumulated score can be determined as the second matching degree based on the seed word benchmark.

[0280] In some more specific embodiments, the seed word benchmark includes the seed word and the first discrimination score corresponding to the seed word. Step 1120 may include, but is not limited to, the following steps:

[0281] Set the counter and initialize it to 0;

[0282] Traversing the word input history records of the object, if a word input history record corresponds to a seed word, accumulating the first discrimination score corresponding to the seed word into a counter;

[0283] After the word input history records of the object are traversed, a second matching degree is determined based on the ratio of the value of the counter to the second threshold.

[0284] It should be noted that the second threshold is a preset threshold, which is used to obtain a ratio with the accumulated first discrimination score to determine the second matching degree. It should be emphasized that the first discrimination score of the seed word is used to quantify the discrimination effect of the seed word. The specific method of obtaining the first discrimination score can refer to the above Figure 6 The corresponding steps are represented by a series of embodiments. Therefore, based on the calculation and accumulation of the first discrimination score of the seed word, and then obtaining a ratio with the second threshold, a reliable and accurate second matching degree can be determined.

[0285] In step 1130, a third degree of matching between the positioning information of the object and the positioning information benchmark is determined. It should be noted that the third degree of matching is used to measure the degree of matching between the positioning information of the object and the positioning information benchmark, and is obtained during the comparison of the positioning information. It should be noted that there are various ways to determine the third degree of matching.

[0286] In some more specific embodiments, you can first request the permission to "query location information" through a pop-up window or jump to a confirmation page, so as to obtain the individual permission or consent of the object. After the individual permission or consent of the target object is clearly obtained, that is, the permission to "query location information" is obtained, the location information of the object can be queried. Among them, if the location information that matches the location information benchmark is queried, the corresponding score is obtained.

[0287] In some more specific embodiments, the positioning information benchmark is a designated area. Based on the positioning information of the object terminal, when the object terminal is within the area, a higher score can be obtained. If the object terminal is at the edge of the area, a medium score can be obtained. If the object terminal is outside the area, the corresponding score is lower.

[0288] In some other more specific embodiments, the positioning information reference may also be the position information of other objects already in the corresponding area. Based on the position information of other objects already in the corresponding area, a position midpoint is obtained, and based on the positioning information of the object terminal, the closer the object terminal is to the position midpoint, the higher the score, and the farther the object terminal is from the position midpoint, the lower the score.

[0289] It should be emphasized that there are various ways to determine the third matching degree, which may include but are not limited to the specific embodiments listed above.

[0290] In step 1140 to step 1150, the degree of matching between the object feature and the object feature benchmark is determined based on the first matching degree, the second matching degree, and the third matching degree. It should be noted that after the first matching degree, the second matching degree, and the third matching degree are determined, the degree of matching between the object feature and the object feature benchmark can be further determined according to the first matching degree, the second matching degree, and the third matching degree, and then based on the matching degree, the object whose object feature matches the object feature benchmark is obtained.

[0291] The embodiment of the present disclosure shown in steps 1110 to 1150 can respectively derive the first degree of match, the second degree of match, and the third degree of match from the three types of comparison processes of terminal usage characteristics, text input habits, and normal activity areas. It can more accurately determine whether the object feature benchmark matches the object feature of the target group based on the first degree of match, the second degree of match, and the third degree of match, thereby improving the grouping accuracy of the target group, generating a more accurate group vocabulary for the target group, and further improving the input efficiency corresponding to the input method.

[0292] Reference Fig.12 The object feature benchmark includes the object terminal application benchmark, the seed word benchmark and the positioning information benchmark. Step 330 may include, but is not limited to, the following steps 1210 to 1250.

[0293] Step 1210, for the single object feature benchmark in the object terminal application benchmark, the seed word benchmark, and the positioning information benchmark, obtain objects whose object features match the single object feature benchmark to form a first group corresponding to the single object feature benchmark;

[0294] Step 1220, if the accuracy of a first group is greater than a third threshold, the first group is used as a target group;

[0295] Step 1230, if the accuracy of each first group is not greater than the third threshold, select any two object feature benchmarks from the object terminal application benchmark, the seed word benchmark, and the positioning information benchmark to form an object feature benchmark combination, and for the object feature benchmark combination, obtain an object whose object feature matches the object feature benchmark combination to form a second group corresponding to the object feature benchmark combination;

[0296] Step 1240, if the accuracy of a second group is greater than a third threshold, use the second group as a target group;

[0297] Step 1250: If the accuracy of each second group is not greater than the third threshold, objects whose object features match the object terminal application benchmark, the seed word benchmark and the positioning information benchmark are obtained to form a target group.

[0298] Steps 1210 to 1250 are described in detail below.

[0299] It should be noted that when the object feature benchmark includes the object terminal application benchmark, the seed word benchmark and the positioning information benchmark; the object feature includes the application installed by the object terminal of the object, the word input history record of the object, and the positioning information of the object. It should be pointed out that if the combination of the object terminal application benchmark, the seed word benchmark and the positioning information benchmark is directly used to determine the members of the target group, a group of relatively small groups will often be framed based on the results of the combination and comparison of multiple benchmark dimensions, so that when the group vocabulary of the target group is subsequently generated, the group vocabulary has a narrower scope of application. Therefore, the improvement of the input efficiency of the input method is not significant. In order to reasonably form a target group according to the comparison process of the application program, the comparison process of the word input history record and the comparison process of the positioning information, the present disclosure also provides some embodiments shown in steps 1210 to 1250.

[0300] In steps 1210 to 1220, for a single object feature benchmark in the object terminal application benchmark, the seed word benchmark, and the positioning information benchmark, objects whose object features match the single object feature benchmark are obtained to form a first group corresponding to the single object feature benchmark. If the accuracy of a first group is greater than a third threshold, the first group is used as a target group. It should be noted that the single object feature benchmark can be any one of the object terminal application benchmark, the seed word benchmark, and the positioning information benchmark. In the embodiment of the present disclosure, it is necessary to first select one from the object terminal application benchmark, the seed word benchmark, and the positioning information benchmark as a single object feature benchmark, and obtain objects whose object features match the single object feature benchmark to form a first group. If the accuracy of a first group is greater than the third threshold, it means that the corresponding single object feature benchmark is sufficient to define the common features of multiple objects in the target group, so the first group with the same object features can be determined as the target group.

[0301] In some more specific embodiments of the present disclosure, step 1220 may further include the following steps: if the accuracy of multiple first groups is greater than a third threshold, the first group with the highest accuracy is used as the target group. In this way, the grouping accuracy of the target group can be improved, a more accurate group word library can be generated for the target group, and the input efficiency corresponding to the input method can be further improved.

[0302] In step 1230 to step 1240, if the accuracy of each first group is not greater than the third threshold, two object feature benchmarks are selected from the object terminal application benchmark, the seed word benchmark and the positioning information benchmark to form an object feature benchmark combination. For the object feature benchmark combination, objects whose object features match the object feature benchmark combination are obtained to form a second group corresponding to the object feature benchmark combination. If the accuracy of a second group is greater than the third threshold, the second group is used as the target group. It should be noted that if the accuracy of each first group is not greater than the third threshold, it means that a single object feature benchmark is not sufficient to define the common features of multiple objects in the target group. Further, it is necessary to select two object feature benchmarks from the object terminal application benchmark, the seed word benchmark and the positioning information benchmark to form an object feature benchmark combination, and obtain objects whose object features match the object feature benchmark combination to form a second group. If the accuracy of a second group is greater than the third threshold, it means that the corresponding object terminal application benchmark is sufficient to define the common features of multiple objects in the target group, so the second group with the same object features can be determined as the target group.

[0303] In some more specific embodiments of the present disclosure, step 1240 may further include the following steps: if the accuracy of multiple second groups is greater than a third threshold, the second group with the greatest accuracy is used as the target group. In this way, the grouping accuracy of the target group can be improved, a more accurate group word library can be generated for the target group, and the input efficiency corresponding to the input method can be further improved.

[0304] Step 1250, if the accuracy of each second group is not greater than the third threshold, obtain objects whose object features match the object terminal application benchmark, the seed word benchmark and the positioning information benchmark to form a target group. It should be noted that if the accuracy of each second group is also not greater than the third threshold, it means that a single object feature benchmark and an object terminal application benchmark are not sufficient to define the common features of multiple objects in the target group. Further, it is necessary to combine the benchmarks of the three dimensions of object features, object terminal application benchmark, seed word benchmark and positioning information benchmark, and obtain objects whose object features match the three dimensions of object terminal application benchmark, seed word benchmark and positioning information benchmark to form a target group.

[0305] In the disclosed embodiment shown by steps 1210 to 1250, a single object feature benchmark is first selected from the object terminal application benchmark, the seed word benchmark, and the positioning information benchmark, and then the single object feature benchmark is used to determine whether it is sufficient to define the common features of multiple objects in the target group; further, an object feature benchmark combination is selected from the object terminal application benchmark, the seed word benchmark, and the positioning information benchmark, and then the object feature benchmark combination is used to determine whether it is sufficient to define the common features of multiple objects in the target group; further, the object terminal application benchmark, the seed word benchmark, and the positioning information benchmark are combined to select objects that match them to form a target group. In this way, the members of the target group can be reasonably determined, which requires that the accuracy needs to be greater than the third threshold, and tries to frame as many people as possible, so that the group vocabulary generated subsequently can be applicable to a larger range.

[0306] Fig.13A -E shows some more specific embodiments of determining the target group.

[0307] like Fig.13A As shown, it corresponds to step 1210 to step 1220. The object set using the input method may specifically include: object a, object b, object c, object d... Further, a single object feature benchmark is selected from the object feature benchmark. Among them, the single object feature benchmark can specifically be any one of the object terminal application benchmark, the seed word benchmark or the positioning information benchmark. After determining the single object feature benchmark, objects whose object features match the single object feature benchmark can be obtained to form a first group corresponding to the single object feature benchmark. Fig.13A In the above, the first group specifically includes: object a, object c, object d, ... If there is a first group whose accuracy is greater than the third threshold, the first group is determined as the target group.

[0308] like Fig. 13B As shown, when the single object feature benchmark is the object terminal application benchmark, the objects matched by the benchmark are grouped into the first group a; when the single object feature benchmark is the seed word benchmark, the objects matched by the benchmark are grouped into the first group b; when the single object feature benchmark is the positioning information benchmark, the objects matched by the benchmark are grouped into the first group c. If the accuracy of the first group a, the first group b, and the first group c are all greater than the third threshold, it means that any one of the object terminal application benchmark, the seed word benchmark, and the positioning information benchmark is sufficient to define the common features of multiple objects in the target group. At this time, the first group with the highest accuracy among the first group a, the first group b, and the first group c can be used as the target group.

[0309] like Fig. 13CAs shown, if the accuracy of each first group is not greater than the third threshold, it is necessary to further select two object feature benchmarks from the object terminal application benchmark, the seed word benchmark and the positioning information benchmark to form an object feature benchmark combination. After determining the object feature benchmark combination, objects whose object features match the object feature benchmark combination can be obtained to form a second group corresponding to the object feature benchmark combination. Fig. 13C In the example, the second group specifically includes: object a, object c, ... If there is a second group whose accuracy is greater than the third threshold, the second group is determined as the target group.

[0310] like Fig.13D As shown, when the object terminal application benchmark is combined with the seed word benchmark to form an object feature benchmark combination, the objects matched by the benchmark are grouped into a second group a; when the seed word benchmark is combined with the positioning information benchmark to form an object feature benchmark combination, the objects matched by the benchmark are grouped into a second group b; when the object terminal application benchmark is combined with the positioning information benchmark to form an object feature benchmark combination, the objects matched by the benchmark are grouped into a second group c. If the accuracy of the second group a, the second group b, and the second group c are all greater than the third threshold, it means that any one of the object terminal application benchmark, the seed word benchmark, and the positioning information benchmark is sufficient to define the common features of multiple objects in the target group. At this time, the second group with the highest accuracy among the second group a, the second group b, and the second group c can be used as the target group.

[0311] like Fig.13E As shown, if the accuracy of each first group is not greater than the third threshold and the accuracy of each second group is not greater than the third threshold, a target group can be formed directly based on objects whose object features match the object terminal application benchmark, seed word benchmark and positioning information benchmark.

[0312] It should be understood that there are various ways to determine the target group, which may include, but not be limited to, the specific embodiments listed above.

[0313] Reference Fig.14 , the accuracy of the first group or the second group is determined by:

[0314] Step 1410, obtaining an evaluation object set, the evaluation object set having a plurality of sample objects, each sample object having a group label to which it belongs;

[0315] Step 1420, obtaining a fourth object belonging to a plurality of sample objects from among the objects in the first group or the second group;

[0316] Step 1430, determining a fifth object whose group label corresponding to the fourth object matches the first group or the second group;

[0317] Step 1440: taking the ratio of the fifth object number to the fourth object number as the accuracy.

[0318] Steps 1410 to 1440 are described in detail below.

[0319] In step 1410, an evaluation object set is obtained, the evaluation object set has a plurality of sample objects, and each sample object has a group label to which it belongs. It should be noted that the group label is used to identify the object group to which the sample object actually belongs.

[0320] In step 1420, a fourth object belonging to multiple sample objects is obtained from the objects in the first group or the second group. It should be noted that after each object in the first group or the second group is identified, it can be queried in the evaluation sample set to determine the fourth object belonging to multiple sample objects.

[0321] In step 1430, a fifth object whose group label corresponding to the fourth object matches the first group or the second group is determined. It should be noted that, since each sample object has a group label to which it belongs, it is only after the fourth object is determined whether the group label of the fourth object matches the first group, or whether the group label of the fourth object matches the second group. In some embodiments of the present disclosure, the object in the first group or the second group that matches the group label corresponding to the fourth object is referred to as the fifth object.

[0322] In step 1440, the ratio of the number of fifth objects to the number of fourth objects is used as the accuracy. It should be noted that, in each fourth object, if the group label is consistent with the first group or the second group, it means that the grouping of this fourth object is accurate, and the number of fifth objects is also the number of accurate groupings. Therefore, using the ratio of the number of fifth objects to the number of fourth objects as the accuracy can reflect the grouping accuracy of the first group, or reflect the grouping accuracy of the second group.

[0323] The embodiment of the present disclosure shown through step 1410 to step 1440 can reasonably and accurately determine the accuracy of the first group or the second group, thereby providing a reference basis for determining the target group members and improving the grouping efficiency of the target group.

[0324] Reference Fig.15A -B, shows some more specific embodiments of determining the accuracy of the first group or the second group.

[0325] like Fig.15AAs shown, the evaluation object set includes sample object a, sample object b, sample object c, sample object d...; wherein, the group label a corresponding to sample object a is University A, the group label b corresponding to sample object b is Company B, the group label c corresponding to sample object c is University A, and the group label d corresponding to sample object d is University A. The first group is the group of University A; it includes the fourth object a belonging to sample object a, the fourth object b belonging to sample object b, and the fourth object c belonging to sample object c, so the number of fourth objects is 3. It should be pointed out that after the evaluation object set and the first group are clarified, it can be clarified that the fifth object whose group label is consistent with the first group includes: the fifth object a whose group label a is consistent with the first group, and the fifth object c whose group label c is consistent with the first group; it can be clarified that the number of fifth objects is 2. Therefore, the ratio of the number of fifth objects to the number of fourth objects is taken as the accuracy; wherein the accuracy can be specifically expressed as: 2 / 3=66.7%.

[0326] like Fig. 15B As shown, the evaluation object set includes sample object a, sample object b, sample object c, sample object d...; wherein, the group label a corresponding to sample object a is University A, the group label b corresponding to sample object b is Company B, the group label c corresponding to sample object c is University A, and the group label d corresponding to sample object d is University A. The second group is the group of University A; it includes the fourth object a belonging to sample object a, the fourth object b belonging to sample object b, the fourth object c belonging to sample object c, and the fourth object d belonging to sample object d, so the number of fourth objects is 4. It should be pointed out that after the evaluation object set and the second group are clarified, it can be clarified that the fifth object whose group label is consistent with the second group includes: the fifth object a whose group label a is consistent with the second group, the fifth object c whose group label c is consistent with the second group, and the fifth object d whose group label d is consistent with the second group; it can be clarified that the number of fifth objects is 3. Therefore, the ratio of the number of fifth objects to the number of fourth objects is taken as the accuracy; wherein the accuracy can be specifically expressed as: 3 / 4=75%.

[0327] It should be understood that there are various ways to determine the accuracy of the first group or the second group, which may include, but not be limited to, the specific embodiments listed above.

[0328] Detailed description of step 340

[0329] Reference Fig.16 , step 340 may include, but is not limited to, the following steps 1610 to 1620.

[0330] Step 1610, for each word in the word input history record, determine a second discrimination score of the word;

[0331] Step 1620: Based on the second discrimination score, determine the target word from the words in the word input history record and add it to the group word library.

[0332] Steps 1610 to 1620 are described in detail below.

[0333] In step 1610 to step 1620, for each word in the word input history record, the second discrimination score of the word is determined; further, based on the second discrimination score, the target word is determined from the words in the word input history record and added to the group vocabulary. It should be noted that the second discrimination score of a word reflects the discriminating role played by the word in the word input history record of the target group. It should be pointed out that after step 330, each object of the target group can be determined, and then for each word in the word input history record of each object of the target group, after the second discrimination score is determined, the word with a discriminating role can be found from the word input history record of each object of the target group according to the second discrimination score, and it can be added as the target word to the group vocabulary, thereby generating or expanding the group vocabulary.

[0334] The embodiment of the present disclosure shown in steps 1610 to 1620 can more accurately and reliably quantify the distinguishing role played by each word in the input history record of the corresponding word of the target group based on the second discrimination score, thereby determining the word with a larger distinguishing role as the target word and adding it to the group vocabulary. In this way, the establishment of the group vocabulary is efficiently completed. When the word that the object wants to enter in the input method has a low frequency of occurrence in the entire domain, but is a word frequently used by objects in a certain field or range, it will fall into the group vocabulary of a certain group. Combining the general vocabulary and the group vocabulary increases the probability of finding the word that the object wants to enter, and improves the input efficiency of the input method.

[0335] Reference Fig.17 , step 1610 may include, but is not limited to, the following steps 1710 to 1730.

[0336] Step 1710, for each word in the word input history record, determining a fourth number of times the word appears in the word input history record;

[0337] Step 1720, obtaining a third number, wherein in the word input history records of each object, the word input history records of a third number of objects contain the word;

[0338] Step 1730, determining a second discrimination score for the word based on the fourth number and the third number.

[0339] Steps 1710 to 1730 are described in detail below.

[0340] In step 1710, for each word in the word input history record, the fourth number of occurrences of the word in the word input history record is determined. It should be noted that the fourth number of occurrences of a word refers to the number of occurrences of the word in the word input history record of the target group.

[0341] In step 1720, a third number is obtained, wherein in the word input history records of each object, the word input history records of a third number of objects contain the word. It should be noted that the third number of a word refers to the number of objects that contain the word in the word input history records of each object in the target group.

[0342] In step 1730, based on the fourth number and the third number, a second discrimination score of the word is determined. In some embodiments, when the fourth number of occurrences of a word in the target group's word input history is higher and the third number of objects that have used the word is smaller, it means that the word has a greater discrimination effect in the target group's word input history. It should be noted that there are various ways to determine the second discrimination score.

[0343] Reference Fig.18 , showing some more specific embodiments for determining the second discrimination score. The word input history records of the target group include: the word input history records of object a, the word input history records of object b, the word input history records of object c, the word input history records of object d, the word input history records of object e, and the word input history records of object f.

[0344] In the word input history of object a, [word A] appears 30 times, [word B] appears 0 times, and [word C] appears 0 times;

[0345] In the word input history of object b, [word A] appears 50 times, [word B] appears 0 times, and [word C] appears 80 times;

[0346] In the word input history of object c, [word A] appears 0 times, [word B] appears 20 times, and [word C] appears 0 times;

[0347] In the word input history of object d, [word A] appears 80 times, [word B] appears 0 times, and [word C] appears 0 times;

[0348] In the word input history of object e, [word A] appears 0 times, [word B] appears 30 times, and [word C] appears 0 times;

[0349] In the word input history of object f, [word A] appears 20 times, [word B] appears 30 times, and [word C] appears 90 times.

[0350] Further, the fourth number of times [Word A], [Word B], and [Word C] appear in the word input history of the target group is determined respectively. Among them, the fourth number of times [Word A] appears is A=30+50+0+80+0+20=180; the fourth number of times [Word B] appears is B=0+0+20+0+30+30=80; the fourth number of times [Word C] appears is 0+80+0+0+0+90=170.

[0351] Furthermore, the third number of objects that have used [word A], [word B], and [word C] is determined. Among them, the objects that have used [word A] are object a, object b, object d, and object f, so the third number A=4; the objects that have used [word B] are object c, object e, and object f, so the third number B=3; the objects that have used [word C] are object b and object f, so the third number c=2.

[0352] Finally, after clarifying the fourth frequency A and the third number A corresponding to [word A], the fourth frequency B and the third number B corresponding to [word B], and the fourth frequency C and the third number C corresponding to [word C], the second discrimination scores corresponding to [word A], [word B], and [word C] can be further determined. Among them, the second discrimination score A corresponding to [word A] = 180 / 4 = 45; the second discrimination score B corresponding to [word B] = 80 / 3 = 26.7; the second discrimination score C corresponding to [word C] = 170 / 2 = 85.

[0353] It should be understood that there are various ways to determine the second discrimination score, which may include, but not be limited to, the specific embodiments listed above.

[0354] Reference Fig.19 , step 1610 may include, but is not limited to, the following steps 1910 to 1920.

[0355] Step 1910, performing word cleaning for each word in the word input history record;

[0356] Step 1920, for each word remaining after word cleaning, determine a second discrimination score of the word.

[0357] Steps 1910 to 1920 are described in detail below.

[0358] In step 1910, word cleaning is performed for each word in the word input history record. It should be noted that word cleaning, also known as word filtering, is a step in text cleaning, which is used to screen out words that are useless for text processing tasks. It should be emphasized that the second discrimination score of a word reflects the distinguishing role played by the word in the word input history record of the target group. It should be understood that not all words in the word input history record need to have a second discrimination score determined. Therefore, it is necessary to perform word cleaning on each word in the word input history record to screen out words that do not need to determine the second discrimination score.

[0359] In some more specific embodiments, the words that can be screened out are called useless words. Among them, there are various types of useless words. In some embodiments, useless words may include function words that often appear in the word input history records and do not express specific meanings. They do not have a distinguishing function and can be removed; in other embodiments, words that do not conform to the input specification of the input method can also be removed. It should be understood that the types of useless words screened out during the word cleaning process may include, but are not limited to the above examples.

[0360] In step 1920, for each word remaining after word cleaning, a second discrimination score of the word is determined. It should be noted that after word cleaning is performed for each word in the word input history record, the screening of useless words is completed immediately, and then the second discrimination score of the word can be further determined for each word remaining after word cleaning.

[0361] In the embodiment of the present disclosure shown by steps 1910 to 1920, word cleaning is performed for each word in the word input history record, and words that do not need to determine the second discrimination score can be screened out, thereby improving the efficiency of determining the second discrimination score, so as to find words with discriminating effect from the word input history records of each object of the target group according to the second discrimination score, and add them as target words to the group vocabulary, thereby generating or expanding the group vocabulary.

[0362] Reference Fig. 20 , step 1910 may include, but is not limited to, the following steps 2010 to 2030.

[0363] Step 2010, identifying the traditional Chinese characters in the words and deleting the words with the traditional Chinese characters;

[0364] Step 220, compare the word with the exclusion word library, and if it is the same as an exclusion word in the exclusion word library, delete the word;

[0365] Step 2030, compare the word with the standard word library, if the word is in the standard word library, retain the word.

[0366] Steps 2010 to 2030 are described in detail below.

[0367] In step 2010, the traditional Chinese characters in the words are identified, and the words with traditional Chinese characters are deleted. According to some embodiments provided by the present disclosure, the input specification of the input method may be the simplified Chinese character specification. If it is recognized that the input text words are mixed with traditional Chinese characters, the traditional Chinese characters therein need to be deleted. It should be noted that the embodiments of the present disclosure can perform the task of identifying traditional Chinese characters through a natural language model or through Optical Character Recognition (OCR). It should be understood that there are various ways to identify traditional Chinese characters, which may include, but are not limited to, the above examples.

[0368] In some more specific embodiments, the input text words include: code path success, fierce code, lamb stewed noodles, smiling always, congratulations; by identifying the traditional Chinese characters in the text words, it can be clearly seen that lamb stewed noodles and smiling always are words with traditional Chinese characters. Therefore, during the process of cleaning words from the word input history record, the words "lamb stewed noodles" and "smiling always" with traditional Chinese characters can be deleted.

[0369] In step 2020, the word is compared with the exclusion word library. If it is the same as an exclusion word in the exclusion word library, the word is deleted. It should be noted that the exclusion word library refers to a pre-set word library for excluding useless words. After comparing the word input history record with the exclusion word library, if there is an exclusion word in the exclusion word library in the word input history record, then the word is deleted during the word cleaning process.

[0370] In some more specific embodiments, the six types of function words, namely adverbs, prepositions, conjunctions, auxiliary words, interjections, and onomatopoeias, often have no specific meaning by themselves. Therefore, they can be deleted during the word cleaning process. Thus, the words included in the exclusion word library can be function words that do not represent specific meanings, such as "yi, ye, wei, suo, qie, er, yin, ze, de, le, me, ne, ba, a". It should be noted that the words included in the exclusion word library can include multiple types, not limited to the specific embodiments listed above.

[0371] In step 2030, the word is compared with the standard word library. If the word is in the standard word library, the word is retained. It should be noted that the standard word library includes multiple words that conform to the word formation norms. It should be understood that the standard word library is used to screen the words in the word input history record that do not contain spelling mistakes. If the word is in the standard word library, it means that the word conforms to the word formation norms, and then the word is retained during the word cleaning process.

[0372] In some more specific embodiments, the input text words include: account number, honey code, book entry law, mutton stewed noodles, smiling often, and congratulations; by comparing the text words with the standard vocabulary, it can be clearly seen that "account number", "honey code" and "book entry law" are words that do not meet the word grouping standards, while "mutton stewed noodles", "smiling often" and "congratulations" are words that meet the word grouping standards, so "mutton stewed noodles", "smiling often" and "congratulations" will be retained during the word cleaning process.

[0373] In some other more specific embodiments, the word combination specifications in the standard word library may include common word combination specifications and custom word combination specifications. Among them, the common word combination specifications include the word combination specifications of common spoken and written language, such as "account", "password" and "input method" which have common meanings; while the custom word combination specifications are suitable for self-made words, for example, although "code success" does not have a common meaning, the word "code success" can be included in the custom word combination specifications, so that the standard word library can have a basis for comparison in the process of comparing with the word input history record.

[0374] It should be understood that the vocabulary included in the standard vocabulary library may include multiple categories and is not limited to the specific embodiments listed above.

[0375] In the embodiment of the present disclosure shown in steps 2010 to 2030, each word in the word input history record is cleaned, and words including traditional Chinese characters can be filtered out, words in the word library are excluded, and words existing in the standard word library are retained, thereby improving the quality of word cleaning. In this way, the efficiency of determining the second discrimination score can be improved, so that words with a discriminating effect can be further found from the word input history records of each object of the target group according to the second discrimination score, and the words are added to the group word library as target words, thereby playing the role of generating or expanding the group word library.

[0376] Reference Fig.21 After step 340, the input method word library generation method further includes:

[0377] Step 2110, obtaining the group name of the target group;

[0378] Step 2120: Generate a group vocabulary name of the group vocabulary based on the group name and a predetermined rule.

[0379] Steps 2110 to 2120 are described in detail below.

[0380] In step 2110, the group name of the target group is obtained. It should be noted that the input method of the embodiment of the present disclosure can be configured with multiple group lexicons, so these group lexicons need to be distinguished by some identification information. Since the object group is a group of objects that meet certain characteristics, the object group often has a group name based on this characteristic. In some embodiments, the group name of the target group can be determined based on the object feature benchmark possessed by the target group. It should be clear that the group name of the target group can be obtained in a variety of ways, which can be obtained through an input device, through network transmission, or through other methods.

[0381] In step 2120, a group word library name of a group word library is generated based on the group name and a predetermined rule. It should be noted that the predetermined rule in the embodiment of the present disclosure refers to a text conversion rule for converting a group name into a group word library name of a group word library.

[0382] In some more specific embodiments, if the group name of the target group is "XX University Group" and the predetermined rule is: uniformly replace the "group" in the group name with "lexicon", then the final group lexicon name generated is "XX University Lexicon"; similarly, if the group name of the target group is "XX Game Player Group" and the predetermined rule is: uniformly delete the "Player Group" in the group name, then the final group lexicon name generated is "XX Game".

[0383] It should be understood that there are various implementations for generating a group vocabulary name, which may include, but not be limited to, the specific embodiments listed above.

[0384] Detailed description of step 2200

[0385] like Fig. 22 As shown, the text input method of an embodiment of the present disclosure may include, but is not limited to, the following steps 2210 to 2240.

[0386] Step 2210, receiving a first encoding string corresponding to a target text input by a target object in an input method interface, wherein the input method has a general word library;

[0387] Step 2220, displaying a group vocabulary control that matches the first code string, wherein the group vocabulary control corresponds to a target group to which the target object belongs;

[0388] Step 2230, in response to triggering the group vocabulary control, obtaining the group vocabulary so as to convert the second encoding string input by the target object into a target text according to the general vocabulary and the group vocabulary; wherein the group vocabulary is generated by the input method vocabulary generation method as described in the above embodiment;

[0389] Step 2240, display the target text.

[0390] Steps 2210 to 2240 are described in detail below.

[0391] In step 2210, in the input method interface, a first code string corresponding to the target text input by the target object is received, and the input method has a general word library. It should be noted that the first code string refers to a code string input by the input device and used to determine the group word library control.

[0392] In some specific embodiments, the first code string may be a code string composed of Chinese, a code string composed of Chinese pinyin, or a code string composed of English. It should be understood that the types of the first code string are various, including but not limited to the above examples.

[0393] In step 2220, the group vocabulary control matching the first coding string is displayed, wherein the group vocabulary control corresponds to the target group to which the target object belongs. It should be noted that after receiving the first coding string corresponding to the target text input by the target object, the group vocabulary control matching the first coding string can be found in the multiple controls of the input method based on the first coding string, and then displayed. It should be emphasized that the group vocabulary control corresponds to the target group to which the target object belongs.

[0394] In step 2230, in response to triggering the group vocabulary control, the group vocabulary is obtained so as to convert the second coded string input by the target object into the target text according to the general vocabulary and the group vocabulary; wherein the group vocabulary is generated by the input method vocabulary generation method of the above embodiment of the present disclosure. It should be noted that the second coded string refers to the coded string input by the input device and used to type the target text. It should be understood that the input method can only obtain the group vocabulary corresponding to the target group after the group vocabulary control is triggered. Therefore, in the process of further using the input method to input the second coded string, the basis for word search includes not only the general vocabulary but also the group vocabulary corresponding to the target group, thereby improving the efficiency of generating the target text.

[0395] In step 2240, the target text is displayed. After the target text is generated, it is displayed on the input method interface, and the input of the target text is completed.

[0396] In the embodiment of the present disclosure shown by steps 2210 to 2240, when the word that the target object wants to input in the input method has a low frequency of occurrence in the whole domain, but is a frequently used word in a certain field or range, it will fall into the group word library corresponding to the target object. Combining the general word library and the group word library of the target object increases the probability of finding the word that the object wants to input, and improves the input efficiency of the input method.

[0397] Reference Fig.23, 2230 may include, but is not limited to, the following steps 2310 to 2320.

[0398] Step 2310, in response to triggering the group vocabulary control, displaying a detail display page of the group vocabulary, the detail display page including a link to download the group vocabulary;

[0399] Step 2320, in response to triggering the link, download the group vocabulary.

[0400] In step 2310 to step 2320, in response to triggering the group vocabulary control, the details display page of the group vocabulary is displayed, and the details display page includes a link to download the group vocabulary; in response to triggering the link, the group vocabulary is downloaded. It should be noted that after the group vocabulary control is triggered, the input method interface will jump to the details display page of the group vocabulary, and the details display page includes a link to download the group vocabulary. Therefore, when the link to download the group vocabulary is triggered, the group vocabulary is immediately downloaded, and the group vocabulary is obtained. Since the details display page includes a link to download the group vocabulary, it is convenient for the target object to download the group vocabulary, so that the convenience of the group vocabulary can be further improved.

[0401] Referring to step 24, step 2220 may include, but is not limited to, the following steps 2410 to 2440.

[0402] Step 2410, obtaining a filtered group vocabulary containing words with the same code as the first code string from the group vocabulary set;

[0403] Step 2420, determining that the target terminal of the target object contains a first application, wherein the first application meets the target terminal application benchmark of the target group corresponding to the filtered group vocabulary;

[0404] Step 2430, determining whether the location information of the target object matches the location information benchmark of the target group corresponding to the filtered group vocabulary;

[0405] Step 2440 , using the filtered group vocabulary as the group vocabulary of the target group, and displaying the group vocabulary control of the group vocabulary.

[0406] Steps 2410 to 2440 are described in detail below.

[0407] In step 2410, a filtered group word library containing words with the same encoding as the first encoding string is obtained from the group word library set. It should be noted that the group word library set includes group word libraries corresponding to multiple object groups. Among the multiple group word libraries in the group word library set, if a group word library has a word with the same encoding as the first encoding string, the word library is determined as the filtered group word library.

[0408] In step 2420, it is determined that the object terminal of the target object contains a first application, wherein the first application meets the object terminal application benchmark of the target group corresponding to the filtered group vocabulary. It should be noted that after determining the filtered group vocabulary, it is necessary to further clarify the object terminal application benchmark of the target group corresponding to the filtered group vocabulary, and it should be pointed out that the application that meets the object terminal application benchmark is the first application. It should be understood that if the object terminal of the target object contains the first application, it means that the target object is more likely to belong to the target group.

[0409] In step 2430, it is determined that the location information of the target object matches the location information benchmark of the target group corresponding to the filtered group vocabulary. It should be noted that after determining the filtered group vocabulary, it is necessary to further clarify the location information benchmark match of the target group corresponding to the filtered group vocabulary. It should be understood that if the location information of the target object matches the location information benchmark of the target group corresponding to the filtered group vocabulary, it means that the target object is more likely to belong to the target group.

[0410] In step 2440, the filtered group word library is used as the group word library of the target group, and the group word library control of the group word library is displayed. It should be noted that after determining that the object terminal of the target object contains the first application, and determining that the location information of the target object matches the location information benchmark of the target group corresponding to the filtered group word library, the possibility that the target object belongs to the target group can be further increased.

[0411] In the embodiment of the present disclosure shown in steps 2410 to 2440, the filtered group vocabulary is used as the group vocabulary of the target group, and the group vocabulary control of the group vocabulary is displayed, so that a more reliable and adaptable group vocabulary can be displayed on the input method interface, thereby improving text input efficiency.

[0412] Overall description of the input method word library generation method and text input method in the disclosed embodiment

[0413] Reference Fig.26 The illustrated embodiment of the present disclosure provides an input method word library generation method, which is intended to generate a group word library corresponding to a target group and applied to an input method.

[0414] Step S1, obtaining the target terminal application benchmark, seed word benchmark and positioning information benchmark;

[0415] Step S2, obtaining an object feature benchmark of the target group according to the target terminal application benchmark, the seed word benchmark and the positioning information benchmark;

[0416] Step S3, obtaining objects whose object features match the object feature benchmark to form a target group;

[0417] It should be noted that the purpose of steps S1 to S3 is to form a target group so as to collect target words from the word input history records of each object in the target group to form a group word library. Specifically, to form a target group, it is necessary to first obtain the object terminal application benchmark, seed word benchmark and positioning information benchmark, and obtain the object feature benchmark of the target group based on the object terminal application benchmark, seed word benchmark and positioning information benchmark, and further obtain objects whose object features match the object feature benchmark to form a target group.

[0418] Step S4: Quality evaluation of the target group;

[0419] It should be noted that the quality evaluation of the target group is intended to evaluate the grouping accuracy of the target group. If the grouping accuracy of the target group is qualified, the quality evaluation passes. If the target group fails the quality evaluation, it is necessary to return to step S2 to re-acquire the object feature benchmark of the target group based on the object terminal application benchmark, the seed word benchmark and the positioning information benchmark, and then perform quality evaluation on the re-acquired target group until the target group passes the quality evaluation.

[0420] Step S5: After passing the quality assessment, word cleaning is performed based on the word input history records of each object in the target group;

[0421] Step S6: determine a target word for each word remaining after word cleaning, and generate a group word library of the target group based on the target word;

[0422] It should be pointed out that through steps S5 to S6, useless words in the input history records of each object word can be screened out. In this way, words with distinguishing functions can be found from the word input history records of each object in the target group, and added to the group vocabulary as target words, thereby playing the role of generating or expanding the group vocabulary.

[0423] Step S7, configuring a group vocabulary name for the group vocabulary;

[0424] Step S8, applying the group vocabulary library configured with the group vocabulary library name to the input method;

[0425] It should be noted that through step S7 to step S8, a group vocabulary name is configured for the group vocabulary, and then the group vocabulary configured with the group vocabulary name can be applied to the input method.

[0426] Reference Fig. 27 The illustrated embodiment of the present disclosure, after generating a group vocabulary corresponding to a target group through an input method vocabulary generation method, further provides a text input method, which aims to use a general vocabulary and a group vocabulary corresponding to a target group to search for input coded strings.

[0427] Step S9: receiving a first code string corresponding to a target text input by a target object in an input method interface, wherein the input method has a general word library;

[0428] Step S10, displaying a group vocabulary control matching the first coding string;

[0429] Step S11, in response to triggering the group vocabulary control, obtaining the group vocabulary, so as to convert the second encoding string input by the target object into a target text according to the general vocabulary and the group vocabulary;

[0430] Step S12: display the target text.

[0431] It should be noted that, through steps S9 to S12, after receiving the first code string corresponding to the target text input by the target object, the group vocabulary control matching the first code string can be determined and displayed. Among them, the group vocabulary control corresponds to the target group to which the target object belongs. Further, after the group vocabulary control is triggered, the group vocabulary corresponding to the target object can be obtained, so as to convert the second code string input by the target object into the target text according to the general vocabulary and the group vocabulary. Further, the target text is displayed, and the input of the target text can be completed.

[0432] In this way, when the subjects of the target group use the input method to input text, they can use the group word library of the target group to search for words. In this way, even if the words input by the subjects have a low frequency of occurrence in the whole domain and are words frequently used by subjects in a certain field or range, they no longer need to be searched and combined word by word in the input method interface, thereby improving the efficiency of text input.

[0433] Description of the apparatus and device of the present disclosure

[0434] It is to be understood that, although the steps in the above-mentioned flowcharts are sequentially displayed according to the characterization of arrows, these steps are not necessarily executed in sequence according to the order of arrow characterization. Unless there is a clear description in the present embodiment, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the above-mentioned flowcharts can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of the steps or stages in other steps.

[0435] It should be noted that in various specific implementations of the present disclosure, when it comes to the need to perform relevant processing based on data related to the characteristics of the target object such as the target object attribute information or attribute information set, the permission or consent of the target object will be obtained first, and the collection, use and processing of these data will comply with relevant laws, regulations and standards. In addition, when the embodiment of the present disclosure needs to obtain the attribute information of the target object, the separate permission or separate consent of the target object will be obtained through a pop-up window or jump to a confirmation page. After clearly obtaining the separate permission or separate consent of the target object, the necessary target object-related data for the normal operation of the embodiment of the present disclosure will be obtained.

[0436] Reference Fig. 27 According to one aspect of the present disclosure, an input method word library generating device 2700 is provided, comprising:

[0437] A first acquisition unit 2710 is used to acquire an object feature reference of a target group;

[0438] The second acquisition unit 2720 is used to acquire the object feature of each object in the object set using the input method;

[0439] The third acquisition unit 2730 is used to acquire objects whose object features match the object feature benchmark in the object set to form a target group;

[0440] The first generating unit 2740 is configured to generate a group word library of the target group based on the word input history records of each object in the target group, for use in the input method.

[0441] Reference Fig.28 According to one aspect of the present disclosure, a text input device 2800 is provided, comprising:

[0442] A first receiving unit 2810 is used to receive a first encoding string corresponding to a target text input by a target object in an input method interface, and the input method has a general word library;

[0443] A first display unit 2820 is used to display a group vocabulary control that matches the first coding string, wherein the group vocabulary control corresponds to a target group to which the target object belongs;

[0444] The fourth acquisition unit 2830 is used to acquire the group vocabulary in response to the triggering of the group vocabulary control, so as to convert the second encoding string input by the target object into the target text according to the general vocabulary and the group vocabulary; wherein the group vocabulary is generated by the input method vocabulary generation method of one aspect of the present disclosure;

[0445] The second display unit 2840 is used to display the target text.

[0446] Optionally, the object feature benchmark includes an object terminal application benchmark, a seed word benchmark, and a positioning information benchmark; the object feature includes an application installed on the object terminal of the object, a word input history record of the object, and the positioning information of the object;

[0447] The third acquisition unit 2730 is specifically used for:

[0448] For an object in the object set, determining that an application installed on an object terminal of the object meets an object terminal application benchmark;

[0449] Determining that the subject's word input history matches the seed word benchmark;

[0450] Determining that the positioning information of the object matches the positioning information reference;

[0451] Add the object to the target group.

[0452] Optionally, the seed word benchmark includes the seed word, and a first discrimination score corresponding to the seed word;

[0453] The third acquisition unit 2730 is specifically used for:

[0454] Set the counter and initialize it to 0;

[0455] Traversing the word input history records of the object, if a word input history record corresponds to a seed word, accumulating the first discrimination score corresponding to the seed word into a counter;

[0456] After traversing the word input history of the object, based on the value of the counter, it is determined that the word input history of the object meets the seed word benchmark.

[0457] Optionally, the first acquisition unit 2710 is specifically used to acquire a seed word, and the seed word is acquired through the following process:

[0458] Acquire a first subgroup and a second subgroup, wherein the first subgroup includes a plurality of first objects that are predetermined to belong to the target group, and the second subgroup includes a plurality of second objects that are predetermined to not belong to the target group;

[0459] Acquire a first corpus, where the first corpus includes a plurality of first sample corpora of first objects;

[0460] Acquire a second corpus, where the second corpus includes a plurality of second sample corpora of second objects;

[0461] Segment each first sample corpus in the first corpus to obtain candidate words;

[0462] For each candidate word, determining a first number of occurrences of the candidate word in the first corpus and a second number of occurrences of the candidate word in the second corpus;

[0463] Based on the first number and the second number, a seed word is determined from the candidate words.

[0464] Optionally, the first obtaining unit 2710 is specifically configured to obtain a first discrimination score, and the first discrimination score is obtained by the following process:

[0465] Acquire a third corpus, where the third corpus includes sample corpora of a first number of third objects;

[0466] Get the third occurrence number of the seed word in the third corpus;

[0467] Obtaining a second number, wherein in the third corpus, sample corpora of a second number of objects contain the seed word, and the second number is not greater than the first number;

[0468] Based on the third number and the second number, a first discrimination score is determined.

[0469] Optionally, the first acquiring unit 2710 is specifically configured to:

[0470] Get the first constant;

[0471] A first discrimination score is determined based on the logarithm of the third number, the first constant, and the logarithm of the second number.

[0472] Optionally, the object feature benchmark includes an object terminal application benchmark, a seed word benchmark, and a positioning information benchmark; the object feature includes an application installed on the object terminal of the object, a word input history record of the object, and the positioning information of the object;

[0473] The third acquisition unit 2730 is specifically used for:

[0474] For an object in the object set, determining a first matching degree between an application installed on an object terminal of the object and an application benchmark of the object terminal;

[0475] Determining a second degree of match between the subject's word input history and the seed word benchmark;

[0476] Determining a third degree of matching between the positioning information of the object and the positioning information benchmark;

[0477] Determining a degree of matching between the object feature and the object feature benchmark based on the first degree of matching, the second degree of matching, and the third degree of matching;

[0478] Based on the matching degree, objects whose object features match the object feature benchmark are obtained.

[0479] Optionally, the seed word benchmark includes the seed word, and a first discrimination score corresponding to the seed word;

[0480] The third acquisition unit 2730 is specifically used for:

[0481] Set the counter and initialize it to 0;

[0482] Traversing the word input history records of the object, if a word input history record corresponds to a seed word, accumulating the first discrimination score corresponding to the seed word into a counter;

[0483] After the word input history records of the object are traversed, a second matching degree is determined based on the ratio of the value of the counter to the second threshold.

[0484] Optionally, the object feature benchmark includes an object terminal application benchmark, a seed word benchmark, and a positioning information benchmark;

[0485] The third acquisition unit 2730 is specifically used for:

[0486] For a single object feature benchmark in the object terminal application benchmark, the seed word benchmark, and the positioning information benchmark, obtaining objects whose object features match the single object feature benchmark to form a first group corresponding to the single object feature benchmark;

[0487] If the accuracy of a first group is greater than a third threshold, taking the first group as a target group;

[0488] If the accuracy of each first group is not greater than the third threshold, select any two object feature benchmarks from the object terminal application benchmark, the seed word benchmark, and the positioning information benchmark to form an object feature benchmark combination, and for the object feature benchmark combination, obtain an object whose object feature matches the object feature benchmark combination to form a second group corresponding to the object feature benchmark combination;

[0489] If the accuracy of a second group is greater than a third threshold, the second group is used as a target group;

[0490] If the accuracy of each second group is not greater than the third threshold, objects whose object features match the object terminal application benchmark, the seed word benchmark and the positioning information benchmark are obtained to form a target group.

[0491] Optionally, the third obtaining unit 2730 is specifically configured to:

[0492] If the accuracy of the plurality of first groups is greater than a third threshold, taking the first group with the greatest accuracy as the target group;

[0493] If the accuracies of the plurality of second groups are greater than the third threshold, the second group with the greatest accuracy is taken as the target group.

[0494] Optionally, the third acquisition unit 2730 is specifically used to determine the accuracy of the first group or the second group, and the accuracy of the first group or the second group is determined in the following manner:

[0495] Obtain an evaluation object set, the evaluation object set has multiple sample objects, and each sample object has a group label to which it belongs;

[0496] Acquire a fourth object belonging to a plurality of sample objects from among the objects in the first group or the second group;

[0497] Determine a fifth object whose group label corresponding to the fourth object matches the first group or the second group;

[0498] The ratio of the fifth object number to the fourth object number is taken as the accuracy.

[0499] Optionally, the first generating unit 2740 is specifically configured to:

[0500] For each word in the word input history, determining a second discrimination score for the word;

[0501] Based on the second discrimination score, a target word is determined from the words in the word input history record and added to the group word library.

[0502] Optionally, the first generating unit 2740 is specifically configured to:

[0503] For each word in the word input history record, determining a fourth number of times the word appears in the word input history record;

[0504] obtaining a third number, wherein, among the word input history records of each object, the word input history records of the third number of objects contain the word;

[0505] Based on the fourth frequency and the third number, a second discrimination score for the word is determined.

[0506] Optionally, the first generating unit 2740 is specifically configured to:

[0507] Perform word cleaning for each word in the word input history record;

[0508] For each word remaining after word cleaning, a second discrimination score for the word is determined.

[0509] Optionally, the first generating unit 2740 is specifically configured to:

[0510] Identify traditional Chinese characters in words and delete words with traditional Chinese characters;

[0511] Compare the word with the exclusion word library, and delete the word if it is the same as an exclusion word in the exclusion word library;

[0512] Compare the word with the standard word library. If the word is in the standard word library, keep the word.

[0513] Optionally, after generating a group word library of the target group based on the word input history records of each object in the target group, the input method word library generating device 2700 further includes:

[0514] a fifth acquiring unit (not shown), configured to acquire a group name of a target group;

[0515] The second generating unit (not shown) generates a group vocabulary name of the group vocabulary based on the group name and a predetermined rule.

[0516] Optionally, the fourth acquiring unit 2830 is specifically configured to:

[0517] In response to triggering the group vocabulary control, displaying a detail display page of the group vocabulary, the detail display page including a link to download the group vocabulary;

[0518] In response to triggering the link, the group vocabulary is downloaded.

[0519] Optionally, the first display unit 2820 is specifically used for:

[0520] Obtaining from the group word library set a filtered group word library containing words with the same code as the first code string;

[0521] Determining that the target terminal of the target object contains a first application, wherein the first application meets the target terminal application benchmark of the target group corresponding to the filtered group vocabulary;

[0522] Determine that the location information of the target object matches the location information benchmark of the target group corresponding to the filtered group vocabulary;

[0523] The filtered group vocabulary is used as the group vocabulary of the target group, and the group vocabulary control of the group vocabulary is displayed.

[0524] Reference Fig.29 , Fig.29 The structural block diagram of the terminal 140 for implementing the text input method of the embodiment of the present disclosure is as follows: the terminal includes: a radio frequency (RF) circuit 2910, a memory 2915, an input unit 2930, a display unit 2940, a sensor 2950, ​​an audio circuit 2960, a wireless fidelity (WiFi) module 2970, a processor 2980, and a power supply 2990. Those skilled in the art will understand that Fig.29 The terminal structure shown in the figure does not constitute a limitation on the mobile phone or computer, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0525] RF circuit 2910 can be used for receiving and sending signals during information transmission or calls. In particular, after receiving the downlink information from the base station, it is sent to processor 2980 for processing; in addition, the designed uplink data is sent to the base station.

[0526] The memory 2915 may be used to store software programs and modules. The processor 2980 executes various functional applications and data processing of the terminal by running the software programs and modules stored in the memory 2915 .

[0527] The input unit 2930 may be used to receive input digital or character information and generate key signal input related to the terminal's settings and function control. Specifically, the input unit 2930 may include a touch panel 2931 and other input devices 2932 .

[0528] The display unit 2940 may be used to display input information or provided information and various menus of the terminal. The display unit 2940 may include a display panel 2941.

[0529] The audio circuit 2960, the speaker 2961, and the microphone 2962 may provide an audio interface.

[0530] In this embodiment, the processor 2980 included in the terminal can execute the text input method of the previous embodiment.

[0531] The terminals of the embodiments of the present disclosure include but are not limited to mobile phones, computers, intelligent voice interaction devices, smart home appliances, vehicle-mounted terminals, aircraft, etc. The embodiments of the present disclosure can be applied to various scenarios, including but not limited to input methods, word processing, etc.

[0532] Reference Fig.30 , Fig.30 The structural block diagram of the part of the server for implementing the input method word library updating method of the embodiment of the present disclosure, the server 110 may have relatively large differences due to different configurations or performances, and may include one or more central processing units (CPU) 3022 (for example, one or more processors) and memory 3032, one or more storage media 3030 (for example, one or more mass storage devices) storing application programs 3042 or data 3044. Among them, the memory 3032 and the storage medium 3030 can be temporary storage or permanent storage. The program stored in the storage medium 3030 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the server 3000. Furthermore, the central processing unit 3022 can be configured to communicate with the storage medium 3030 and execute a series of instruction operations in the storage medium 3030 on the server 3000.

[0533] The server 3000 may also include one or more power supplies 3026, one or more wired or wireless network interfaces 3050, one or more input and output interfaces 3058, and / or one or more operating systems 3041, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, etc.

[0534] The processor in the server 3000 can be used to execute the input method vocabulary updating method of the embodiment of the present disclosure.

[0535] The embodiments of the present disclosure further provide a computer-readable storage medium, which is used to store program codes, and the program codes are used to execute the input method vocabulary updating method or text input method of the aforementioned embodiments.

[0536] The embodiment of the present disclosure also provides a computer program product, which includes a computer program. A processor of a computer device reads and executes the computer program, so that the computer device executes the above-mentioned input method word library update method or text input method.

[0537] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present disclosure described herein can, for example, be implemented in an order other than those illustrated or described herein. In addition, the terms "comprises" and "comprising" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0538] It should be understood that in the present disclosure, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0539] It should be understood that in the description of the embodiments of the present disclosure, the meaning of multiple (or multiple items) is more than two, greater than, less than, exceed, etc. are understood to not include the number, and above, below, within, etc. are understood to include the number.

[0540] In the several embodiments provided in the present disclosure, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0541] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0542] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0543] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present disclosure is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the various embodiments of the present disclosure. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store program codes.

[0544] It should also be understood that the various implementations provided in the embodiments of the present disclosure can be combined arbitrarily to achieve different technical effects.

[0545] The above is a specific description of the implementation methods of the present disclosure, but the present disclosure is not limited to the above implementation methods. Technical personnel familiar with the art can also make various equivalent modifications or substitutions without violating the spirit of the present disclosure. These equivalent modifications or substitutions are all included in the scope defined by the claims of the present disclosure.

Claims

1. A method for generating an input method vocabulary, characterized in that: include: Obtaining object feature benchmarks for the target group; Obtaining object features of each object in an object set using the input method; In the object set, obtaining objects whose object features match the object feature benchmark to form the target group; Based on the word input history records of each of the objects in the target group, a group word library of the target group is generated for use in the input method.

2. The input method word library generation method according to claim 1, characterized in that: The object feature benchmark includes an object terminal application benchmark, a seed word benchmark and a positioning information benchmark; the object feature includes an application installed on the object terminal of the object, a word input history of the object, and the positioning information of the object; The step of acquiring objects in the object set whose object features match the object feature reference to form the target group includes: For the object in the object set, determining that an application installed on an object terminal of the object meets the object terminal application benchmark; Determining that the subject's word input history matches the seed word benchmark; Determining that the positioning information of the object matches the positioning information reference; The subject is added to the target group.

3. The input method word library generation method according to claim 2, characterized in that: The seed word benchmark includes a seed word and a first discrimination score corresponding to the seed word; The determining that the word input history of the object meets the seed word benchmark includes: Set a counter, the counter is initialized to 0; Traversing the word input history records of the object, if one of the word input history records corresponds to a seed word, accumulating the first discrimination score corresponding to the seed word into the counter; After traversing the word input history record of the object, based on the value of the counter, it is determined that the word input history record of the object meets the seed word benchmark.

4. The input method word library generation method according to claim 3, characterized in that: The seed words are obtained through the following process: Acquire a first seed group and a second seed group, wherein the first seed group includes a plurality of first objects that are predetermined to belong to the target group, and the second seed group includes a plurality of second objects that are predetermined to not belong to the target group; Acquire a first corpus, where the first corpus includes a plurality of first sample corpora of the first objects; Acquire a second corpus, where the second corpus includes a plurality of second sample corpora of the second object; Segment each first sample corpus in the first corpus to obtain candidate words; For each of the candidate words, determining a first number of occurrences of the candidate word in the first corpus and a second number of occurrences of the candidate word in the second corpus; The seed word is determined from the candidate words based on the first number of times and the second number of times.

5. The input method word library generation method according to claim 3, characterized in that: The first discrimination score is obtained by the following process: Acquire a third corpus, wherein the third corpus includes sample corpora of a first number of third objects; Obtain the third number of times the seed word appears in the third corpus; Acquire a second number, wherein in the third corpus, sample corpora of the second number of objects contain the seed word, and the second number is not greater than the first number; Based on the third number and the second number, the first discrimination score is determined.

6. The input method word library generation method according to claim 5, characterized in that: The determining the first discrimination score based on the third number and the second number comprises: Get the first constant; The first discrimination score is determined based on the logarithm of the third number, the first constant, and the logarithm of the second number.

7. The input method word library generation method according to claim 1, characterized in that: The object feature benchmark includes an object terminal application benchmark, a seed word benchmark and a positioning information benchmark; the object feature includes an application installed on the object terminal of the object, a word input history of the object, and the positioning information of the object; The step of acquiring objects in the object set whose object features match the object feature reference to form the target group includes: For the object in the object set, determining a first matching degree between an application installed on an object terminal of the object and an application benchmark of the object terminal; determining a second degree of match between the subject's word input history record and the seed word benchmark; Determining a third degree of matching between the positioning information of the object and the positioning information benchmark; Determining a degree of match between the object feature and the object feature benchmark based on the first degree of match, the second degree of match, and the third degree of match; Based on the matching degree, an object whose object feature matches the object feature reference is obtained.

8. The input method word library generation method according to claim 1, characterized in that: The object feature benchmark includes an object terminal application benchmark, a seed word benchmark and a positioning information benchmark; The step of acquiring objects in the object set whose object features match the object feature reference to form the target group includes: For the object terminal application benchmark, the seed word benchmark, and a single object feature benchmark in the positioning information benchmark, obtaining an object whose object feature matches the single object feature benchmark to form a first group corresponding to the single object feature benchmark; If the accuracy of one of the first groups is greater than a third threshold, taking the first group as the target group; If the accuracy of each of the first groups is not greater than a third threshold, select any two object feature benchmarks from the object terminal application benchmark, the seed word benchmark, and the positioning information benchmark to form an object feature benchmark combination, and for the object feature benchmark combination, obtain objects whose object features match the object feature benchmark combination to form a second group corresponding to the object feature benchmark combination; If the accuracy of one of the second groups is greater than the third threshold, taking the second group as the target group; If the accuracy of each of the second groups is not greater than the third threshold, objects whose object features match the object terminal application benchmark, the seed word benchmark and the positioning information benchmark are obtained to form the target group.

9. The input method word library generation method according to claim 8, characterized in that: The accuracy of the first group or the second group is determined by: Acquire an evaluation object set, wherein the evaluation object set has a plurality of sample objects, and each of the sample objects has a group label to which it belongs; Acquire a fourth object belonging to a plurality of the sample objects from among the objects in the first group or the second group; Determine a fifth object whose group label corresponding to the fourth object matches the first group or the second group; The ratio of the fifth number of objects to the fourth number of objects is used as the accuracy.

10. The input method word library generation method according to claim 1, characterized in that: The generating a group word library of the target group based on the word input history records of each of the objects in the target group includes: For each word in the word input history, determining a second discrimination score for the word; Based on the second discrimination score, a target word is determined from the words in the word input history record and added to the group word library.

11. The input method word library generation method according to claim 10, characterized in that: The step of determining, for each word in the word input history record, a second discrimination score of the word comprises: For each word in the word input history record, word cleaning is performed; For each of the words remaining after word cleaning, a second discrimination score for the word is determined.

12. The input method word library generation method according to claim 11, characterized in that: The word cleaning is performed for each word in the word input history record, including: Identify the traditional Chinese characters in the words, and delete the words containing the traditional Chinese characters; Compare the word with an exclusion word library, and if the word is identical to an exclusion word in the exclusion word library, delete the word; The word is compared with a standard word library, and if the word is in the standard word library, the word is retained.

13. The input method word library generation method according to claim 1, characterized in that: After generating a group word library of the target group based on the word input history records of each of the objects in the target group, the input method word library generation method further includes: Obtaining the group name of the target group; Based on the group name and a predetermined rule, a group vocabulary name of the group vocabulary is generated.

14. A text input method, characterized in that: include: In an input method interface, receiving a first encoding string corresponding to a target text input by a target object, the input method having a universal word library; Displaying a group vocabulary control matching the first encoding string, wherein the group vocabulary control corresponds to the target group to which the target object belongs; In response to triggering the group vocabulary control, a group vocabulary is obtained so as to convert the second encoding string input by the target object into the target text according to the general vocabulary and the group vocabulary; wherein the group vocabulary is generated by the input method vocabulary generation method according to any one of claims 1 to 13; The target text is displayed.

15. The text input method according to claim 14, characterized in that: The acquiring of the group vocabulary in response to triggering the group vocabulary control includes: In response to triggering the group vocabulary control, displaying a detail display page of the group vocabulary, wherein the detail display page includes a link for downloading the group vocabulary; In response to triggering the link, the group vocabulary is downloaded.

16. An input method word library generating device, characterized in that: include: A first acquisition unit, used to acquire an object feature reference of a target group; A second acquisition unit, used to acquire an object feature of each object in an object set using the input method; A third acquisition unit, configured to acquire objects in the object set that match the object feature with the object feature reference to form the target group; The first generating unit is configured to generate a group word library of the target group based on the word input history records of each of the objects in the target group, for use in the input method.

17. A text input device, characterized in that: include: A first receiving unit, used for receiving a first encoding string corresponding to a target text input by a target object in an input method interface, wherein the input method has a universal word library; A first display unit, configured to display a group vocabulary control matching the first coding string, wherein the group vocabulary control corresponds to the target group to which the target object belongs; a fourth acquisition unit, configured to acquire a group vocabulary in response to triggering the group vocabulary control, so as to convert the second encoding string input by the target object into the target text according to the general vocabulary and the group vocabulary; wherein the group vocabulary is generated by the input method vocabulary generation method according to any one of claims 1 to 13; The second display unit is used to display the target text.

18. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, it implements the input method vocabulary generation method described in any one of claims 1 to 13, or implements the text input method described in any one of claims 14 to 15.

19. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, it implements the input method vocabulary generation method described in any one of claims 1 to 13, or implements the text input method described in any one of claims 14 to 15.

20. A computer program product, comprising a computer program, wherein the computer program is read and executed by a processor of a computer device, so that the computer device executes the input method vocabulary generation method described in any one of claims 1 to 13, or executes the text input method described in any one of claims 14 to 15.

Images