A word combination method, device, electronic device and readable storage medium

By identifying and ignoring irrelevant basic words in intelligent word grouping technology and generating word grouping paths that do not contain secondary nodes, the problem of word grouping results deviation in the existing technology is solved, and the accuracy of word grouping results is improved and User input efficiency.

CN110807321BActive Publication Date: 2024-11-12BEIJING SOGOU TECHNOLOGY DEVELOPMENT CO LTD
View PDF 9 Cites 0 Cited by

Patent Information

Application Number
CN201810806457.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2018-07-20
Publication Date
2024-11-12
Estimated Expiration
2039-08-04

AI Technical Summary

Technical Problem

The accuracy of word grouping results in existing intelligent word grouping technology is low, which affects the user's input efficiency. Especially when there are irrelevant words in the sentences entered by the user, it leads to a deviation in word grouping results, and the user needs to manually form words.

Method used

By obtaining the word grouping path corresponding to the string entered by the user, find the basic words that meet the preset conditions as secondary nodes, generate a second word group path that does not contain secondary nodes, and determine the target word group path, thereby improving the Accuracy of word grouping results.

Benefits of technology

It effectively improves the accuracy of word grouping results, reduces the user's need to manually form words, and improves input efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN110807321B_ABST
    Figure CN110807321B_ABST
Patent Text Reader

Abstract

The present invention provides a word grouping method, device, electronic device and readable storage medium, the word grouping method comprising: obtaining a plurality of first word grouping paths corresponding to a character string input by a user, each first word grouping path including a plurality of basic words; using the basic words satisfying a preset condition in each first word grouping path as the secondary node of the first word grouping path; generating a second word grouping path based on the other basic words in each first word grouping path except the secondary node; determining a target word grouping path according to the generated second word grouping path, and obtaining a word grouping result based on the first word grouping path corresponding to the target word grouping path. The technical solution provided by the present invention can effectively improve the accuracy of the word grouping result, thereby improving the input efficiency of the user.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of input methods, and in particular to a word formation method, device, electronic device and readable storage medium. Background Art

[0002] With the continuous development of science and technology, input methods have been rapidly developed, and users have higher and higher requirements for the ease of use of input methods. Therefore, intelligent word combination technology is applied in input methods to improve the input efficiency of users.

[0003] The existing intelligent word grouping technology obtains the optimal word grouping path based on the word network and ngram model, and then obtains the word grouping result based on the optimal word grouping path. However, the accuracy of the word grouping results obtained based on the existing intelligent word grouping technology is still low. In many cases, users still need to manually group words, which affects the user's input efficiency. Summary of the invention

[0004] In view of the above problems, the present application provides a word formation method, device, electronic device and readable storage medium, which are used to solve the technical problems in the prior art that the accuracy of word formation results is low and affects the user input efficiency.

[0005] In a first aspect, an embodiment of the present invention provides a word grouping method, the method comprising: obtaining a plurality of first word grouping paths corresponding to a character string input by a user, each of the first word grouping paths comprising a plurality of basic words; using the basic words in each of the first word grouping paths that meet preset conditions as secondary nodes of the first word grouping paths; generating a second word grouping path based on other basic words in each of the first word grouping paths except the secondary nodes; determining a target word grouping path based on the generated second word grouping path, and obtaining a word grouping result based on the first word grouping path corresponding to the target word grouping path.

[0006] Optionally, the method of using the basic words that meet the preset conditions in each of the first-group word paths as the secondary nodes of the first-group word paths includes: searching for basic words in each of the first-group word paths that do not have a preset N-gram relationship with adjacent preset basic words, and using the found basic words as the secondary nodes of the first-group word paths, wherein the preset N-gram relationship is a binary relationship or a relationship higher than binary.

[0007] Optionally, the method of using the basic words that meet preset conditions in each of the first group of words paths as the secondary nodes of the first group of words paths includes: matching each basic word in each of the first group of words paths with the feature words in a preset feature dictionary, and determining the successfully matched basic words as the secondary nodes of the first group of words paths, wherein the feature dictionary includes multiple feature words, and the feature dictionary is pre-obtained based on a preset corpus and a preset binary library.

[0008] Optionally, the preset corpus includes a plurality of historical word grouping data manually triggered by a user.

[0009] Optionally, before matching each basic word in the first group of word paths with a feature word in a preset feature dictionary, the method further includes: receiving and storing the feature dictionary sent by a server.

[0010] Optionally, after determining the target word grouping path based on the generated second word grouping path, and obtaining the word grouping result based on the first word grouping path corresponding to the target word grouping path, it also includes: detecting whether the upper screen result triggered by the user is the word grouping result; when it is not the word grouping result, adding the upper screen result to the preset corpus.

[0011] Optionally, matching each basic word in each of the first group of word paths with feature words in a preset feature dictionary includes: for each basic word in each of the first group of word paths, determining whether there is a feature word identical to the basic word in the feature dictionary; if so, determining that the basic word is successfully matched with the feature word in the feature dictionary.

[0012] Optionally, matching each basic word in each of the first group of word paths with feature words in a preset feature dictionary includes: for each basic word in each of the first group of word paths, determining whether there is a feature word in the feature dictionary whose similarity with the basic word is greater than a preset similarity threshold; if so, determining that the base word is successfully matched with the feature word in the feature dictionary.

[0013] Optionally, determining the target word group path based on the generated second word group path includes: obtaining a score of each of the generated second word group paths according to a preset algorithm; and taking the second word group path with the highest score as the target word group path.

[0014] Optionally, obtaining the word grouping result based on the first word grouping path corresponding to the target word grouping path includes: concatenating basic words in the first word grouping path corresponding to the target word grouping path to obtain the word grouping result.

[0015] In a second aspect, an embodiment of the present invention further provides a word grouping device, the device comprising: a first acquisition module, a first determination module, a generation module and a second determination module. The first acquisition module is used to acquire a plurality of first word grouping paths corresponding to a character string input by a user, each of the first word grouping paths comprising a plurality of basic words. The first determination module is used to use the basic words that meet preset conditions in each of the first word grouping paths as secondary nodes of the first word grouping path. The generation module is used to generate a second word grouping path based on other basic words in each of the first word grouping paths except the secondary nodes. The second determination module is used to determine a target word grouping path based on the generated second word grouping path, and obtain a word grouping result based on the first word grouping path corresponding to the target word grouping path.

[0016] Optionally, the first determination module is specifically used to: search for basic words in each of the first group of word paths that do not have a preset N-gram relationship with adjacent preset basic words, and use the found basic words as secondary nodes of the first group of word paths, wherein the preset N-gram relationship is a binary relationship or a relationship higher than binary.

[0017] Optionally, the first determination module is specifically used to: match each basic word in each of the first group of word paths with the feature words in a preset feature dictionary, and determine the successfully matched basic words as the secondary nodes of the first group of word paths, wherein the feature dictionary includes multiple feature words, and the feature dictionary is pre-obtained based on a preset corpus and a preset binary library.

[0018] Optionally, the preset corpus includes a plurality of historical word grouping data manually triggered by a user.

[0019] Optionally, the device further includes: a second acquisition module, configured to receive and store the feature dictionary sent by a server.

[0020] Optionally, the device further comprises: a detection module, configured to detect whether the on-screen result triggered by the user is the word combination result; when it is not the word combination result, adding the on-screen result to the preset corpus.

[0021] Optionally, the first determination module is specifically used to: for each basic word in each of the first group of word paths, determine whether there is a feature word identical to the basic word in the feature dictionary; if so, determine that the basic word successfully matches the feature word in the feature dictionary.

[0022] Optionally, the first determination module is specifically used to: for each basic word in each of the first group of word paths, determine whether there is a feature word in the feature dictionary whose similarity with the basic word is greater than a preset similarity threshold; if so, determine that the basic word successfully matches the feature word in the feature dictionary.

[0023] Optionally, the second determination module is specifically used to: obtain a score of each of the generated second word group paths according to a preset algorithm; and use the second word group path with the highest score as the target word group path.

[0024] Optionally, the second determination module is specifically configured to: concatenate basic words in the first word grouping path corresponding to the target word grouping path to obtain the word grouping result.

[0025] In a third aspect, an embodiment of the present invention further provides an electronic device, including a processor and a memory, wherein the memory is coupled to the processor. The memory stores instructions, and when the instructions are executed by the processor, the electronic device performs the following operations: obtaining a plurality of first word group paths corresponding to a character string input by a user, each of which includes a plurality of basic words; using the basic words that meet the preset conditions in each of the first word group paths as the secondary nodes of the first word group path; generating a second word group path based on the other basic words in each of the first word group paths except the secondary nodes; determining a target word group path based on the generated second word group path, and obtaining a word group result based on the first word group path corresponding to the target word group path.

[0026] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the word formation method provided in the first aspect above.

[0027] In the technical solution of the embodiment of the present invention, the first word group path corresponding to the character string input by the user is first obtained, and each first word group path includes multiple basic words. Then, the basic words that meet the preset conditions in each first word group path are found as the secondary nodes of the first word group path. These secondary nodes will seriously affect the word group probability of the first word group path. Thereafter, based on the other basic words in each first word group path except the secondary nodes, a second word group path is generated, and then the target word group path is determined according to the generated second word group path, and the word group result is obtained based on the first word group path corresponding to the target word group path. By ignoring the secondary nodes on the first word group path in the process of determining the target word group path, the deviation of the word group result caused by the secondary nodes is avoided, and the accuracy of the word group result can be effectively improved, thereby improving the input efficiency of the user.

[0028] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0030] Figure 1 A method flow chart of a word grouping method provided for the first embodiment of the present invention;

[0031] Figure 2 A functional module block diagram of a word grouping device provided in the second embodiment of the present invention;

[0032] Figure 3 A schematic diagram of the structure of an electronic device according to an exemplary embodiment of the present invention;

[0033] Figure 4 A schematic diagram of the structure of an electronic device used as a server according to a third embodiment of the present invention. DETAILED DESCRIPTION

[0034] Based on the problem that the accuracy of word grouping results in existing input methods is not high, which affects the user's input efficiency, the inventor of this application has found through long-term research that in some cases, the sentence that the user wants to input is to insert one or more words between two originally related words, which will cause these words to be unrelated, resulting in deviations in the word grouping results, and the word grouping results that the user wants cannot be obtained. The user needs to manually group words, which affects the user's input efficiency. For example, the user wants to input "cut hair once", and the corresponding input string is "liyicifa", where "li" and "fa" have a binary relationship, but there is no binary relationship between "li" and "once" and between "once" and "fa", then according to the existing intelligent word grouping technology, "cut hair once" will not be grouped out. For another example, the user wants to input "storage is very efficient", but there is no binary relationship between "storage" and "very" and between "very" and "efficient", then "storage is very efficient" will not be grouped out.

[0035] Based on this, the embodiments of the present invention provide a word combination method, which can effectively improve the accuracy of word combination results, thereby improving the input efficiency of users. The word combination method provided by the embodiments of the present invention is applied to an electronic device. The electronic device may be a user terminal installed with an input method, such as a personal computer (PC), a mobile phone, a tablet computer, a wearable device, a personal digital assistant, etc. Alternatively, the electronic device may also be a server.

[0036] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. The components of the embodiments of the present invention described and illustrated herein usually can be arranged and designed in various different configurations. Therefore, the detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely represents the selected embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0037] As Figure 1 shown, the first embodiment of the present invention provides a word combination method. The word combination method includes:

[0038] Step S101, obtaining a plurality of first word combination paths corresponding to the string input by the user, and each of the first word combination paths includes a plurality of basic words;

[0039] Among them, the string can be encoded input under various input methods, such as pinyin input, Wubi input, handwriting input, etc. After the input method receives the string input by the user, it will segment the string, and then construct a word network by querying the word library to obtain one or more word combination paths. In this embodiment, for the convenience of distinction, the word combination paths obtained according to the word network are called the first word combination paths, and each of the first word combination paths includes a plurality of basic words arranged in sequence.

[0040] Taking the Chinese pinyin input method as an example, for example, the string input by the user is "liyicifa", and the syllable segmentation is first performed as li’yi’ci’fa. For the convenience of explanation, assume that the constructed word network is: li{Li, Reason}, yi’ci{Once}, fa{Send}, so that the first word combination path A and the first word combination path B can be obtained. The first word combination path A is "Li - Once - Send", and the first word combination path B is "Reason - Once - Send". Among them, the first word combination path A includes three basic words arranged in sequence, namely "Li", "Once" and "Send", and the first word combination path B also includes three basic words arranged in sequence, namely "Reason", "Once" and "Send".

[0041] It can be understood that in the above two first group of word paths, there is no binary relationship between "Li" and "once", "Li" and "once", and "once" and "hair". If words are grouped directly according to these two first group of word paths, since the word frequency of "Li" is higher than that of "Li", the word grouping result will be "Li once hair", which is obviously not the word grouping result desired by the user, that is, the word grouping result deviates. Therefore, after performing the above step S101, the following step S102 is continuously executed for each of the obtained first group of word paths to obtain the secondary nodes in each first group of word paths, that is, the basic words that cause the deviation of the word grouping result. For example, the basic word "once" in the first group of word path B.

[0042] Step S102: Use the basic words that meet the preset conditions in each of the first group of word paths as the secondary nodes of the first group of word paths;

[0043] In an embodiment of the present invention, the secondary node in the first group of word paths is the basic word that does not have a preset N - ary relationship with the adjacent preset basic word among the basic words of the first group of word paths. That is to say, the preset condition in step S102 is that there is no preset N - ary relationship with the adjacent preset basic word. At this time, the above step S102 can specifically be: find the basic words that do not have a preset N - ary relationship with the adjacent preset basic word among the basic words of each of the first group of word paths, and use the found basic words as the secondary nodes of the first group of word paths.

[0044] In this embodiment, the preset N - ary relationship can be a binary relationship or a relationship of more than two elements, that is, N is an integer greater than or equal to 2, and can be specifically set according to actual needs. Among them, the binary relationship, also known as 2 - gram, is used to represent the probability of two elements appearing successively. The relationship of more than two elements is used to represent the probability of more than two elements appearing successively. For example, the ternary relationship can be used to represent the probability of the words "hold down", "shift", and "key" appearing successively. Corresponding to different settings of the preset N - ary relationship, the above step S102 also has various implementation manners.

[0045] In an embodiment of the present invention, the above - mentioned preset N - ary relationship is a binary relationship. At this time, the above - mentioned preset basic word can include the previous basic word and / or the next basic word. Correspondingly, the secondary node in the first group of word paths can be: the basic word that does not have a binary relationship with both the adjacent previous basic word and the next basic word among the basic words of the first group of word paths. Or, the secondary node in the first group of word paths can be: the basic word that does not have a binary relationship with the adjacent previous basic word among the basic words of the first group of word paths. Or, the secondary node in the first group of word paths can be: the basic word that does not have a binary relationship with the adjacent next basic word among the basic words of the first group of word paths.

[0046] Specifically, when the preset N-ary relationship is a binary relationship, the above step S102 has multiple implementation methods.

[0047] First, a feature dictionary may be pre-set, and the feature dictionary includes multiple feature words. The above step S102 includes: matching each basic word in each first group word path with a feature word in a preset feature dictionary, and determining the successfully matched basic word as a secondary node of the first group word path.

[0048] Among them, the feature dictionary can be obtained in advance based on a preset corpus and a preset binary library. At this time, the basic word that successfully matches the feature word in the feature dictionary is the basic word that does not meet the above-mentioned preset conditions. When the word grouping method provided in this embodiment is applied to a user terminal, the feature dictionary can be obtained by the server based on a preset corpus and a pre-stored binary library and sent to the user terminal. At this time, before each basic word in the first word grouping path is matched with the feature word in the preset feature dictionary, the above step S102 also includes: receiving and storing the feature dictionary sent by the server. Alternatively, the above feature dictionary can also be obtained in advance locally on the user terminal based on a preset corpus and a preset binary library. At this time, the above step S102 also includes: obtaining the feature dictionary based on the preset corpus and the preset binary library.

[0049] Among them, the preset corpus may include a plurality of historical word combination data manually triggered by users. Specifically, user logs may be counted in advance to find out a plurality of historical word combination data that trigger word combination and are manually triggered by users as the preset corpus. It is understandable that the user will manually trigger word combination when the intelligent word combination result cannot obtain the result the user wants. Using the historical word combination data manually triggered by users to construct a corpus is conducive to quickly and accurately obtaining some feature words that affect the results of intelligent word combination and generating a feature dictionary. Of course, in other embodiments of the present invention, the preset corpus may also be other corpora, such as an Internet corpus obtained based on web crawler technology or a corpus accumulated by a cloud computing input method.

[0050] The preset binary library can be obtained by pre-statisticing the binary relationship of the required vocabulary based on the historical corpus. For example, the required vocabulary may include but is not limited to: Chinese / English vocabulary, English / English vocabulary, Chinese / Chinese vocabulary, etc. Among them, the historical corpus can be an Internet corpus obtained based on web crawler technology, or it can be a corpus accumulated by cloud computing input method. Among them, the Internet can be an Internet blog corpus, an Internet news corpus and / or an Internet forum corpus, etc., and the embodiment of the present invention does not limit the specific historical corpus. Alternatively, the binary library can be pre-acquired in other ways and stored in the user terminal. A plurality of word pairs with binary relationships are stored in the binary library, and each word pair includes two words. Among them, the two words can include Chinese vocabulary and Chinese vocabulary, Chinese vocabulary and English vocabulary, English vocabulary and Chinese vocabulary, and English vocabulary and English vocabulary, etc.

[0051] When the preset corpus includes multiple historical word combination data manually triggered by users, the step of obtaining a feature dictionary based on the preset corpus and the preset binary library may specifically include: performing word segmentation processing on each historical word combination data in the preset corpus, and obtaining multiple word segments constituting the historical word combination data in turn; obtaining multiple feature words based on the preset binary library and each word segment of each of the historical word combination data, and constituting the feature dictionary based on the multiple feature words.

[0052] Specifically, the step of obtaining multiple feature words based on the preset binary library and each word segment of each of the historical word group data may include: based on the preset binary library, obtaining the word segment that does not have a binary relationship with the adjacent preset word segment in each word segment of each of the historical word group data as the target word segment; and obtaining the multiple feature words based on the obtained target word segment.

[0053] The preset participle may include a previous participle and a next participle, or may be the previous participle, or may be the next participle, and may be set as needed in actual application.

[0054] When the preset participle includes the previous participle and the next participle, based on the preset binary library, the participles in each historical word group data can be searched for participles that have no binary relationship with the adjacent previous participle and the next participle, and the searched participles are all used as target participles.

[0055] When the preset segmentation word is the previous segmentation word, the segmentation words that do not have a binary relationship with the adjacent previous segmentation word in each segmentation word of each historical word group data can be searched based on the preset binary library, and the searched segmentation words are all used as target segmentations.

[0056] When the preset segmentation word is the next segmentation word, the segmentation words that do not have a binary relationship with the next adjacent segmentation word in each historical word group data can be searched based on the preset binary library, and the searched segmentation words are all used as target segmentations.

[0057] It should be noted that in the above process, the method for determining whether there is a binary relationship between two participles can be: determine whether the word pair consisting of the two participles is in a preset binary library. If the word pair is not in the binary library, it is determined that there is no binary relationship between the two participles. If the word pair is in the binary library, it is determined that there is a binary relationship between the two participles.

[0058] For example, a certain historical word group data includes the following segmentations: F1, F2 and F3. When the preset segmentation includes the previous segmentation and the next segmentation, the previous segmentation adjacent to F2 is F1, and the next segmentation adjacent to F2 is F3. If the word pair {F1, F2} and the word pair {F2, F3} are not in the binary library, that is, there is no binary relationship between F1 and F2 and between F2 and F3, then F2 is used as the target segmentation. When the preset segmentation is the previous segmentation, for segmentation F2, the previous segmentation adjacent to F2 is F1. If the word pair {F1, F2} is not in the binary library, that is, there is no binary relationship between F1 and F2, then F2 is used as the target segmentation. When the preset segmentation word is the next segmentation word, for segmentation word F1, the next segmentation word adjacent to F1 is F2. If the word pair {F1, F2} is not in the binary library, that is, there is no binary relationship between F1 and F2, F1 is used as the target segmentation word.

[0059] In an embodiment of the present invention, obtaining the plurality of feature words according to the acquired target participles may include: acquiring the frequency of each of the target participles, and taking the target participles whose frequency exceeds a preset frequency threshold as the feature words.

[0060] For example, the acquired target participles can be added to a preset list. If the target participle already exists in the list, the frequency of the target participle is increased by 1, thereby obtaining the frequency of each target participle. The frequency of each target participle is then compared with the preset frequency threshold, and the target participles whose frequency exceeds the preset frequency threshold are all used as the feature words, thereby obtaining a feature dictionary. Among them, the preset frequency threshold can be set according to actual needs, for example, it can be set to 5, 10 or 20, etc. This can effectively improve the accuracy of the feature words, thereby improving the accuracy of the secondary node search results, and then improving the accuracy of the word group results.

[0061] Of course, in other embodiments of the present invention, all the acquired target participles may be used as feature words.

[0062] In addition, according to different requirements, the above implementation methods of matching each basic word in the first group of word paths with the feature words in the preset feature dictionary may include the following two implementation methods:

[0063] As a first implementation, for each basic word in each of the first word group paths, it can be determined whether there is a feature word identical to the basic word in the feature dictionary, and if so, it is determined that the basic word successfully matches the feature word in the feature dictionary. For example, for the first word group path B, "once" is in the feature dictionary, then it is determined that "once" successfully matches the feature word in the feature dictionary, that is, "once" is a secondary node of the first word group path B.

[0064] As a second implementation, for each basic word in each of the first group of word paths, it can be determined whether there is a feature word in the feature dictionary whose similarity with the basic word is greater than a preset similarity threshold. If so, it is determined that the basic word successfully matches the feature word in the feature dictionary.

[0065] Among them, the preset similarity threshold can be set as needed. Specifically, the user terminal can pre-store the word vector of each feature word in the feature dictionary. Based on the word vector corresponding to each basic word in each first group of word paths and the word vector of each feature word, the similarity between each basic word and each feature word in each first group of word paths is obtained. When there is a similarity greater than the preset similarity threshold in the similarity corresponding to each basic word, it is determined that the basic word matches the feature word in the feature dictionary successfully, that is, the basic word is the secondary node of the corresponding first group of word paths. For example, for the "once" in the above-mentioned first group of word paths B, if the feature dictionary includes the feature word "twice", the similarity between "once" and "twice" is greater than the preset similarity threshold, then it is determined that "once" matches the feature word in the feature dictionary successfully, that is, "once" is the secondary node of the first group of word paths B.

[0066] The above method of searching for secondary nodes in the first word grouping path based on a preset feature dictionary can effectively improve the search efficiency of the secondary nodes, thereby improving the word grouping efficiency on the basis of improving the accuracy of the word grouping results.

[0067] Of course, in addition to the above-mentioned method, the embodiment of the present invention also provides a second implementation method of step S102, which is specifically: detecting whether the corresponding basic word of each first group of word paths has a binary relationship with the adjacent preset basic word, and taking the basic word that does not have a binary relationship with the adjacent preset basic word as the secondary node of the first group of word paths. For example, when the above-mentioned preset basic word includes the previous basic word and the next basic word, the above-mentioned step S102 may specifically include:

[0068] Step S201, traverse all first group word paths, and take any first group word path as the current first group word path;

[0069] Step S202, taking all basic words except the first basic word and the last basic word in the current first group of word paths as target basic words of the current first group of word paths;

[0070] Step S203, traverse the target basic words of the current first group of word paths, take any target basic word of the current first group of word paths as the current basic word, and perform the following detection steps:

[0071] Determine whether there is a binary relationship between the previous basic word adjacent to the current basic word in the current first group of word paths and the current basic word, and determine whether there is a binary relationship between the current basic word and the next basic word adjacent to the current basic word in the current first group of word paths; if there is no binary relationship between the previous basic word adjacent to the current basic word and the current basic word, and between the current basic word and the next basic word adjacent to the current basic word, then determine that the current basic word meets the preset conditions, and use the current basic word as a secondary node of the current first group of word paths; if there is a binary relationship between the previous basic word adjacent to the current basic word and the current basic word, or between the current basic word and the next basic word adjacent to the current basic word, then determine that the current basic word does not meet the preset conditions, that is, the current basic word is not a secondary node of the current first group of word paths;

[0072] Step S204, taking the next target basic word of the current first group of word paths as the current basic word, and repeating the above detection steps until all target basic words of the current first group of word paths are detected;

[0073] Step S205 , taking the next first group word path as the current first group word path, and repeatedly executing steps S202 to S205 until the search for each first group word path obtained in the above step S101 is completed.

[0074] For example, the basic words included in the current first group of word paths are: P1, P2, P3 and P4, then P2 and P3 are both the target basic words of the current first group of word paths. Take P2 as the current basic word. Obviously, the previous basic word adjacent to P2 is P1, and the next basic word adjacent to P2 is P3. If the word pair {P1, P2} and the word pair {P2, P3} are not in the preset binary library, it means that there is no binary relationship between P1 and P2 and between P2 and P3, then P2 is the secondary node of the current first group of word paths. Then take P3 as the current basic word, and judge whether P3 is a secondary node according to the above method.

[0075] Of course, in other embodiments of the present invention, the above-mentioned preset basic word may also be the previous basic word, or, it may also be the next basic word. Accordingly, when the above-mentioned preset basic word is the previous basic word, the secondary node of the first group of word paths is: among the basic words of the first group of word paths, the basic word that does not have a binary relationship with the adjacent previous basic word. When the above-mentioned preset basic word is the next basic word, the secondary node of the first group of word paths is: among the basic words of the first group of word paths, the basic word that does not have a binary relationship with the adjacent next basic word. It can be understood that when the above-mentioned preset basic word is the previous basic word or the next basic word, the specific search process of the secondary node of the first group of word paths is similar to the implementation method when the above-mentioned preset basic word is the previous basic word and the next basic word, and will not be described in detail here.

[0076] In addition, in one embodiment of the present invention, the above-mentioned preset N-gram relationship can also be a ternary relationship. In this case, the above-mentioned preset basic words include: the first two basic words and / or the last two basic words. That is, the secondary node in the first group of word paths is the basic word in each basic word of the first group of word paths that does not have a ternary relationship with the adjacent first two basic words and / or the last two basic words. It can be understood that at this time, the specific search process of the secondary node of the first group of word paths is similar to the implementation method when the above-mentioned preset N-gram relationship is a binary relationship, and will not be described in detail here.

[0077] Of course, in other embodiments of the present invention, the above-mentioned preset N-ary relationship may also be a relationship of ternary or more (not including ternary), which may be specifically set as needed.

[0078] Step S103, generating a second group of word paths based on other basic words in each of the first group of word paths except the secondary nodes;

[0079] After obtaining the secondary nodes of each first group of word paths based on the above step S102, for each first group of word paths, other basic words except the secondary nodes are obtained according to the arrangement order of the basic words in the first group of word paths as the main nodes of the first group of word paths, and the second group of word paths corresponding to the first group of word paths are formed based on the obtained main nodes. For example, in the above example, the secondary node of the first group of word paths is P2, and the main nodes of the first group of word paths are: P1, P3 and P4 in sequence, therefore, the basic words included in the second group of word paths corresponding to the first group of word paths are: P1, P3 and P4 in sequence.

[0080] In addition, it can be understood that, for some first group of word paths without secondary nodes, the generated second group of word paths is the same as the first group of word paths.

[0081] Step S104, determining a target word grouping path according to the generated second word grouping path, and obtaining a word grouping result based on the first word grouping path corresponding to the target word grouping path.

[0082] Determining the target word grouping path according to the second word grouping path generated in step S103 can avoid the influence of the secondary nodes on the word grouping result, thereby improving the accuracy of the word grouping result. Specifically, the step of determining the target word grouping path according to the generated second word grouping path may include: obtaining the score of each of the generated second word grouping paths according to a preset algorithm; and taking the second word grouping path whose score meets the preset score condition as the target word grouping path.

[0083] For example, a preset ngram model may be used to calculate the conditional probability of each second group of word paths, and the calculated conditional probability is used as the score of the second group of word paths.

[0084] In addition, the preset score condition can be set according to actual needs. In this embodiment, the second word group path with the highest score can be used as the target word group path. In other embodiments of the present invention, a score threshold can also be set as needed, and the second word group paths with scores exceeding the score threshold can be used as target word group paths. Alternatively, the scores of all second word group paths are sorted, and the second word group paths with scores ranked in the front preset number of places are used as target word group paths, and when displaying the candidate items, the word group results corresponding to the target word group paths are displayed in order from large to small in score. Among them, the preset number of places can be set according to actual needs.

[0085] Specifically, the process of obtaining the word grouping result based on the first word grouping path corresponding to the target word grouping path specifically includes: splicing the basic words in the first word grouping path corresponding to the target word grouping path to obtain the word grouping result. That is to say, after determining the target word grouping path, all the basic words on the first word grouping path corresponding to the target word grouping path, including the main nodes and the secondary nodes, are spliced ​​in order to obtain the word grouping result.

[0086] For example, for the above example, when the character string input by the user is "liyicifa", the first word path A is obtained: "Li-yishi-fa" and the first word path B is obtained: "Li-yishi-fa". After the above step S102, it is determined that "yishi" in the first word path A and the second word path B are both secondary nodes, so the second word path A' corresponding to the first word path A is obtained: "Li-fa", and the second word path B' corresponding to the first word path B is obtained: "Li-fa". Obviously, the score of the second word path B' is higher than the score of the second word path A'. When the second word path with the highest score is used as the target word path, the second word path B' is used as the target word path, so that all the basic words on the first word path B are spliced ​​in order to obtain the word result "Li-yishifa". In this way, the word result that the user wants can be obtained, and the user does not need to manually compose words.

[0087] In addition, in one embodiment of the present invention, when the above-mentioned step S102 specifically adopts a method of searching for secondary nodes in the first word group path based on a preset feature dictionary, after executing the above-mentioned step S104, the word grouping method may also include: detecting whether the upper screen result triggered by the user is the said word grouping result; when it is not the said word grouping result, adding the said upper screen result to the said preset corpus.

[0088] It is understandable that after obtaining the word combination result corresponding to the character string input by the user according to the above steps S101 to S104, the input method will display the word combination result as a candidate item, and the user will perform the on-screen operation based on the candidate item. When the candidate item is the word combination result that the user wants, the user can directly select the candidate item to be on the screen, that is, the on-screen result is the candidate item. When the candidate item is not the word combination result that the user wants, the user will perform a manual word combination operation to obtain the desired on-screen result. Therefore, when it is detected that the on-screen result triggered by the user is not the word combination result obtained by step S104, it indicates that the on-screen result is manually triggered by the user. Then, the on-screen result is added to the preset corpus to learn the user's personalized data and dynamically update the feature dictionary, which is conducive to improving the accuracy of the secondary node search results, thereby further improving the accuracy of the word combination results.

[0089] It should be noted that when it is detected that the screen result triggered by the user is the word combination result obtained in step S104, or when no user triggering screen operation is detected, the above operation of adding the screen result to the preset corpus is not performed.

[0090] The word grouping method provided by the embodiment of the present invention distinguishes between the main nodes and the secondary nodes on the first word grouping path. In the process of determining the target word grouping path, the secondary nodes on the first word grouping path are ignored and only the main nodes are considered, thereby avoiding the deviation of the word grouping results caused by the secondary nodes, and can effectively improve the accuracy of the word grouping results, thereby improving the input efficiency of the user.

[0091] Please refer to Figure 2 , shows a module block diagram of a word grouping device provided by the second embodiment of the present invention. The word grouping device 200 includes:

[0092] A first acquisition module 201 is used to acquire a plurality of first word paths corresponding to a character string input by a user, each of which includes a plurality of basic words;

[0093] A first determination module 202, configured to use the basic words satisfying the preset conditions in each of the first word group paths as the secondary nodes of the first word group paths;

[0094] A generating module 203, configured to generate a second group of word paths based on other basic words in each of the first group of word paths except the secondary nodes;

[0095] The second determination module 204 is used to determine a target word grouping path according to the generated second word grouping path, and obtain a word grouping result based on the first word grouping path corresponding to the target word grouping path.

[0096] As an optional implementation, the first determination module 202 is specifically used to: search for basic words in each of the first group of word paths that do not have a preset N-gram relationship with adjacent preset basic words, and use the found basic words as secondary nodes of the first group of word paths, wherein the preset N-gram relationship is a binary relationship or a relationship higher than binary.

[0097] As an optional implementation manner, the preset N-ary relationship is a binary relationship, and the preset basic words include: a previous basic word and / or a next basic word.

[0098] As an optional implementation, the first determination module 202 is specifically used to: match each basic word in each of the first group of word paths with the feature words in a preset feature dictionary, and determine the successfully matched basic words as the secondary nodes of the first group of word paths, wherein the feature dictionary includes multiple feature words, and the feature dictionary is pre-obtained based on a preset corpus and a preset binary library.

[0099] As an optional implementation, the preset corpus includes a plurality of historical word grouping data manually triggered by a user.

[0100] As an optional implementation, the word combination device 200 further includes: a second acquisition module, which is used to receive and store the feature dictionary sent by the server when the word combination device 200 provided by this embodiment runs on the user terminal.

[0101] As an optional implementation, the word grouping device 200 further includes: a detection module, configured to detect whether the on-screen result triggered by the user is the word grouping result; when it is not the word grouping result, adding the on-screen result to the preset corpus.

[0102] As an optional implementation, the first determination module 202 is specifically used to: for each basic word in each of the first group of word paths, determine whether there is a feature word identical to the basic word in the feature dictionary; if so, determine that the basic word successfully matches the feature word in the feature dictionary.

[0103] As an optional implementation, the first determination module 202 is specifically used to: for each basic word in each of the first group of word paths, determine whether there is a feature word in the feature dictionary whose similarity with the basic word is greater than a preset similarity threshold; if so, determine that the basic word is successfully matched with the feature word in the feature dictionary.

[0104] As an optional implementation, the second determination module 204 is specifically used to: obtain the score of each of the second word group paths generated according to a preset algorithm; and use the second word group path with the highest score as the target word group path.

[0105] As an optional implementation, the second determining module 204 is specifically configured to: concatenate the basic words in the first word grouping path corresponding to the target word grouping path to obtain the word grouping result.

[0106] It should be noted that the specific manner in which each module of the word grouping device 200 provided in the embodiment of the present invention performs operations has been described in detail in the embodiment of the method, and will not be elaborated on in detail here.

[0107] In addition, a third embodiment of the present invention provides an electronic device, including a processor and a memory, wherein the memory is coupled to the processor, and the memory stores instructions, and when the instructions are executed by the processor, the electronic device performs the following operations:

[0108] Acquire a plurality of first word group paths corresponding to the character string input by the user, each of the first word group paths including a plurality of basic words;

[0109] Taking the basic words that meet the preset conditions in each of the first group of word paths as the secondary nodes of the first group of word paths;

[0110] Based on other basic words except the secondary nodes in each of the first group of word paths, generating a second group of word paths;

[0111] A target word grouping path is determined according to the generated second word grouping path, and a word grouping result is obtained based on the first word grouping path corresponding to the target word grouping path.

[0112] It should be noted that, in the electronic device provided in the embodiment of the present invention, the specific implementation of each of the above steps and the technical effects produced are the same as those in the aforementioned method embodiment. For the sake of brief description, the parts not mentioned in this embodiment can be referred to the corresponding contents in the aforementioned method embodiment.

[0113] Figure 3 8 is a block diagram of an electronic device 800 according to an exemplary embodiment. For example, the electronic device 800 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0114] Please refer to Figure 3 , the electronic device 800 may include one or more of the following components: a processing component 802 , a memory 804 , a power component 806 , a multimedia component 808 , an audio component 810 , an input / output (I / O) interface 812 , a sensor component 814 , and a communication component 816 .

[0115] The processing component 802 generally controls the overall operation of the electronic device 800, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 802 may include one or more processors 820 to execute instructions to complete all or part of the steps of the word grouping method provided in the first embodiment above. In addition, the processing component 802 may include one or more modules to facilitate the interaction between the processing component 802 and other components. For example, the processing component 802 may include a multimedia module to facilitate the interaction between the multimedia component 808 and the processing component 802.

[0116] The memory 804 is configured to store various types of data to support operations on the device 800. Examples of such data include instructions for any application or method operating on the electronic device 800, contact data, phone book data, messages, pictures, videos, etc. The memory 804 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0117] The power component 806 provides power to the various components of the electronic device 800. The power component 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the electronic device 800.

[0118] The multimedia component 808 includes a screen that provides an output interface between the electronic device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 808 includes a front camera and / or a rear camera. When the device 800 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera may receive external multimedia data. Each front camera and the rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

[0119] The audio component 810 is configured to output and / or input audio signals. For example, the audio component 810 includes a microphone (MIC), and when the electronic device 800 is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 804 or sent via the communication component 816. In some embodiments, the audio component 810 also includes a speaker for outputting audio signals.

[0120] I / O interface 812 provides an interface between processing component 802 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include but are not limited to: a home button, a volume button, a start button, and a lock button.

[0121] The sensor assembly 814 includes one or more sensors for providing various aspects of status assessment for the electronic device 800. For example, the sensor assembly 814 can detect the open / closed state of the device 800, the relative positioning of components, such as the display and keypad of the electronic device 800, and the sensor assembly 814 can also detect the position change of the electronic device 800 or a component of the electronic device 800, the presence or absence of contact between the user and the electronic device 800, the orientation or acceleration / deceleration of the electronic device 800, and the temperature change of the electronic device 800. The sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 814 may also include an optical sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 814 may also include an accelerometer, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0122] The communication component 816 is configured to facilitate wired or wireless communication between the electronic device 800 and other devices. The electronic device 800 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 816 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 816 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0123] In an exemplary embodiment, the electronic device 800 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above methods.

[0124] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 804 including instructions, and the instructions can be executed by a processor 820 of an electronic device 800 to perform the above method. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0125] A non-temporary computer-readable storage medium, when the instructions in the storage medium are executed by a processor of an electronic device, enables the electronic device to execute a word grouping method, the method comprising: obtaining a plurality of first word grouping paths corresponding to a character string input by a user, each of the first word grouping paths comprising a plurality of basic words; using the basic words that meet preset conditions in each of the first word grouping paths as secondary nodes of the first word grouping paths; generating a second word grouping path based on other basic words in each of the first word grouping paths except the secondary nodes; determining a target word grouping path based on the generated second word grouping path, and obtaining a word grouping result based on the first word grouping path corresponding to the target word grouping path.

[0126] Figure 4 1 is a schematic diagram of the structure when the electronic device is a server in an embodiment of the present invention. The server 1900 may have relatively large differences due to different configurations or performances, and may include one or more central processing units (CPU) 1922 (for example, one or more processors) and memory 1932, and one or more storage media 1930 (for example, one or more mass storage devices) storing application programs 1942 or data 1944. Among them, the memory 1932 and the storage medium 1930 can be short-term storage or permanent storage. The program stored in the storage medium 1930 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations on the server. Furthermore, the central processing unit 1922 can be configured to communicate with the storage medium 1930 and execute a series of instruction operations in the storage medium 1930 on the server 1900.

[0127] The server 1900 may also include one or more power supplies 1926, one or more wired or wireless network interfaces 1950, one or more input and output interfaces 1958, one or more keyboards 1956, and / or, one or more operating systems 1941, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, etc.

[0128] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art that are not disclosed in this disclosure. The specification and examples are to be considered exemplary only, and the true scope and spirit of the present invention are indicated by the following claims.

[0129] It should be understood that the present invention is not limited to the precise structure described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

[0130] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A word formation method, characterized in that: The method comprises: Acquire a plurality of first word group paths corresponding to the character string input by the user, each of the first word group paths including a plurality of basic words; Taking the basic words that meet the preset conditions in each of the first group of word paths as the secondary nodes of the first group of word paths; Based on other basic words except the secondary nodes in each of the first group of word paths, generating a second group of word paths; A target word grouping path is determined according to the generated second word grouping path, and a word grouping result is obtained based on the first word grouping path corresponding to the target word grouping path.

2. The method according to claim 1, characterized in that The step of using the basic words that meet the preset conditions in each of the first word group paths as the secondary nodes of the first word group paths includes: Search for basic words in each of the first group of word paths that do not have a preset N-ary relationship with adjacent preset basic words, and use the found basic words as secondary nodes of the first group of word paths, wherein the preset N-ary relationship is a binary relationship or a relationship of more than two elements.

3. The method according to claim 1, characterized in that The step of using the basic words that meet the preset conditions in each of the first word group paths as the secondary nodes of the first word group paths includes: Each basic word in each of the first group of word paths is matched with a feature word in a preset feature dictionary, and the successfully matched basic word is determined as a secondary node of the first group of word paths, wherein the feature dictionary includes multiple feature words, and the feature dictionary is pre-obtained based on a preset corpus and a preset binary library.

4. The method according to claim 3, characterized in that The preset corpus includes a plurality of historical word grouping data manually triggered by users.

5. The method according to claim 3, characterized in that: Before matching each basic word in each of the first word paths with a feature word in a preset feature dictionary, the method further includes: Receive and store the feature dictionary sent by the server.

6. The method according to claim 3, characterized in that After determining the target word grouping path according to the generated second word grouping path, and obtaining the word grouping result based on the first word grouping path corresponding to the target word grouping path, the method further includes: Detecting whether the on-screen result triggered by the user is the word combination result; When it is not the word group result, the on-screen result is added to the preset corpus.

7. The method according to claim 3, characterized in that The step of matching each basic word in each of the first word group paths with a feature word in a preset feature dictionary comprises: For each basic word in each of the first group of word paths, it is determined whether there is a feature word identical to the basic word in the feature dictionary. If so, it is determined that the basic word successfully matches the feature word in the feature dictionary.

8. The method according to claim 3, characterized in that The step of matching each basic word in each of the first word group paths with a feature word in a preset feature dictionary comprises: For each basic word in each of the first group of word paths, it is determined whether there is a feature word in the feature dictionary whose similarity with the basic word is greater than a preset similarity threshold; if so, it is determined that the basic word successfully matches the feature word in the feature dictionary.

9. The method according to claim 1, characterized in that: Determining the target word grouping path according to the generated second word grouping path includes: Obtaining a score for each of the second group of word paths generated according to a preset algorithm; The second word group path with the highest score is used as the target word group path.

10. The method according to claim 1, characterized in that The obtaining of the word grouping result based on the first word grouping path corresponding to the target word grouping path includes: The basic words in the first word grouping path corresponding to the target word grouping path are concatenated to obtain the word grouping result.

11. A word forming device, characterized in that: The device comprises: A first acquisition module, used to acquire a plurality of first word group paths corresponding to the character string input by the user, each of the first word group paths including a plurality of basic words; A first determination module, configured to use the basic words satisfying the preset conditions in each of the first word group paths as the secondary nodes of the first word group paths; A generating module, configured to generate a second group of word paths based on other basic words in each of the first group of word paths except the secondary nodes; The second determination module is used to determine a target word grouping path according to the generated second word grouping path, and obtain a word grouping result based on the first word grouping path corresponding to the target word grouping path.

12. The device according to claim 11, characterized in that The first determining module is specifically used for: Search for basic words in each of the first group of word paths that do not have a preset N-ary relationship with adjacent preset basic words, and use the found basic words as secondary nodes of the first group of word paths, wherein the preset N-ary relationship is a binary relationship or a relationship of more than two elements.

13. The device according to claim 11, characterized in that The first determining module is specifically used for: Each basic word in each of the first group of word paths is matched with a feature word in a preset feature dictionary, and the successfully matched basic word is determined as a secondary node of the first group of word paths, wherein the feature dictionary includes multiple feature words, and the feature dictionary is pre-obtained based on a preset corpus and a preset binary library.

14. The device according to claim 13, characterized in that The preset corpus includes a plurality of historical word grouping data manually triggered by users.

15. The device according to claim 13, characterized in that The device also includes: The second acquisition module is used to receive and store the feature dictionary sent by the server.

16. The device according to claim 13, characterized in that The device also includes: The detection module is used to detect whether the on-screen result triggered by the user is the word combination result; when it is not the word combination result, the on-screen result is added to the preset corpus.

17. The device according to claim 13, characterized in that The first determining module is specifically used for: For each basic word in each of the first group of word paths, it is determined whether there is a feature word identical to the basic word in the feature dictionary. If so, it is determined that the basic word successfully matches the feature word in the feature dictionary.

18. The device according to claim 13, characterized in that The first determining module is specifically used for: For each basic word in each of the first group of word paths, it is determined whether there is a feature word in the feature dictionary whose similarity with the basic word is greater than a preset similarity threshold; if so, it is determined that the basic word successfully matches the feature word in the feature dictionary.

19. The device according to claim 11, characterized in that The second determining module is specifically used for: Obtaining a score for each of the second group of word paths generated according to a preset algorithm; The second word group path with the highest score is used as the target word group path.

20. The device according to claim 11, characterized in that The second determination module is specifically used to: splice the basic words in the first word grouping path corresponding to the target word grouping path to obtain the word grouping result.

21. An electronic device, characterized in that: The electronic device includes a processor and a memory, wherein the memory is coupled to the processor and stores instructions, and when the instructions are executed by the processor, the electronic device causes the electronic device to perform the following operations: Acquire a plurality of first word group paths corresponding to the character string input by the user, each of the first word group paths including a plurality of basic words; Taking the basic words that meet the preset conditions in each of the first group of word paths as the secondary nodes of the first group of word paths; Based on other basic words except the secondary nodes in each of the first group of word paths, generating a second group of word paths; A target word grouping path is determined according to the generated second word grouping path, and a word grouping result is obtained based on the first word grouping path corresponding to the target word grouping path.

22. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method described in any one of claims 1 to 10 are implemented.

Citation Information

Patent Citations

  • Chinese auto-answer method and system

    CN101178711A

  • Encoded string sequencing method, device and character input method and device

    CN101644961A

  • Phrase recommendation method and apparatus

    CN105138512A

  • Wrong word knowledge base construction method based on fuzzy matching and statistics

    CN105512110A

  • Algorithm for globally optimizing quality of text keywords

    CN106598941A