CN113705221A - Word pushing method and device, electronic equipment and storage medium - Google Patents
Word pushing method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN113705221A CN113705221A CN202110995589.3A CN202110995589A CN113705221A CN 113705221 A CN113705221 A CN 113705221A CN 202110995589 A CN202110995589 A CN 202110995589A CN 113705221 A CN113705221 A CN 113705221A
- Authority
- CN
- China
- Prior art keywords
- word
- initial
- words
- user
- vector
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 76
- 239000013598 vector Substances 0.000 claims abstract description 114
- 238000013136 deep learning model Methods 0.000 claims description 18
- 238000004364 calculation method Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 11
- 238000012163 sequencing technique Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 239000000126 substance Substances 0.000 claims description 2
- 230000008569 process Effects 0.000 description 15
- 238000004891 communication Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 8
- 238000012545 processing Methods 0.000 description 7
- 210000004556 brain Anatomy 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000003993 interaction Effects 0.000 description 2
- 235000013372 meat Nutrition 0.000 description 2
- 230000009182 swimming Effects 0.000 description 2
- 235000013311 vegetables Nutrition 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000003446 memory effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Abstract
The disclosure provides a word pushing method and device, electronic equipment and a storage medium, and relates to the technical field of internet application, in particular to the technical field of language learning. The specific implementation scheme is as follows: acquiring a word vector of an initial word currently pushed to a user; respectively calculating the similarity between each word in the word library and the initial word according to the word vector of the initial word and the word vector of each word in the word library, wherein the word library comprises a plurality of words and word vectors of the plurality of words; determining adjacent words corresponding to the initial words based on the similarity; and pushing the adjacent words corresponding to the initial words to the user. The efficiency of memorizing words of the user is improved.
Description
Technical Field
The present disclosure relates to the field of language learning technologies, and in particular, to a word pushing method and apparatus, an electronic device, and a storage medium.
Background
In the process of memorizing words, people easily spend a lot of time but are difficult to obtain a good memory effect. With the development of internet technology, more and more online word memorizing software appears in the market, and aims to help users to effectively memorize words by utilizing fragmented time and achieve the best word memorizing effect while not wasting time.
Disclosure of Invention
The disclosure provides a word pushing method, a word pushing device, an electronic device and a storage medium.
According to an aspect of the present disclosure, there is provided a word pushing method, including:
acquiring a word vector of an initial word currently pushed to a user;
respectively calculating the similarity between each word in the word library and the initial word according to the word vector of the initial word and the word vector of each word in the word library, wherein the word library comprises a plurality of words and word vectors of the plurality of words;
determining adjacent words corresponding to the initial words based on the similarity;
and pushing the adjacent words corresponding to the initial words to the user.
According to another aspect of the present disclosure, there is provided a word pushing apparatus including:
the word vector acquisition module is used for acquiring a word vector of an initial word currently pushed to a user;
the similarity calculation module is used for respectively calculating the similarity between each word in the word library and the initial word according to the word vector of the initial word and the word vector of each word in the word library, wherein the word library comprises a plurality of words and word vectors of the plurality of words;
the adjacent word determining module is used for determining the adjacent word corresponding to the initial word based on each similarity;
and the adjacent word pushing module is used for pushing the adjacent word corresponding to the initial word to the user.
According to another aspect of the present disclosure, there is provided an electronic device including:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform any of the word push methods described above.
According to another aspect of the present disclosure, there is provided a non-transitory computer readable storage medium storing computer instructions for causing a computer to perform any of the word push methods described above.
According to another aspect of the present disclosure, there is provided a computer program product comprising a computer program which, when executed by a processor, implements any of the word push methods described above.
The word pushing method comprises the steps of firstly obtaining a word vector of an initial word currently pushed to a user, then respectively calculating the similarity between each word in a word library and the initial word according to the word vector of the initial word and the word vector of each word in the word library, then determining an adjacent word corresponding to the initial word based on each similarity, and pushing the adjacent word corresponding to the initial word to the user.
Therefore, by the word pushing method provided by the disclosure, words are pushed to the user in the learning process of the user.
It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present disclosure, nor do they limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.
Drawings
The drawings are included to provide a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:
fig. 1 is a schematic flow chart diagram of a first embodiment of a word push method provided according to the present disclosure;
FIG. 2 is a flowchart illustrating a second embodiment of a word push method provided in accordance with the present disclosure;
FIG. 3 is a flowchart illustrating a third embodiment of a word push method according to the present disclosure;
FIG. 4 is a flowchart illustrating one possible implementation manner of step S13 in the word push method provided according to the present disclosure;
FIG. 5 is a schematic flow chart diagram of a fourth embodiment of a word push method provided in accordance with the present disclosure;
FIG. 6 is a schematic flow chart diagram of a fifth embodiment of a word push method provided in accordance with the present disclosure;
FIG. 7 is a schematic structural diagram of a word pushing device provided according to the present disclosure;
FIG. 8 is a block diagram of an electronic device for implementing the word push method of an embodiment of the present disclosure.
Detailed Description
Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, in which various details of the embodiments of the disclosure are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
In the related technology, when the online word memory software pushes words to the user, an Evonghaus memory curve model is often utilized to follow the human brain memory law and push the words to the user in a periodic review mode to continuously deepen the memory of the user for the words. However, since the number of words is large, the user is helped to memorize the words only by the human brain memory law, which makes the efficiency of memorizing the words of the user still low.
In order to solve the problem, the present disclosure provides a word pushing method, an apparatus, an electronic device, and a storage medium, where the word pushing method in the related art is applied to help a user memorize words, and only depends on the human brain memory law, so that the efficiency of the user in memorizing words is low, and the word pushing method includes:
acquiring a word vector of an initial word currently pushed to a user;
respectively calculating the similarity between each word in the word library and the initial word according to the word vector of the initial word and the word vector of each word in the word library, wherein the word library comprises a plurality of words and word vectors of the plurality of words;
determining adjacent words corresponding to the initial words based on the similarity;
and pushing the adjacent words corresponding to the initial words to the user.
Therefore, by applying the word pushing method provided by the disclosure, the similarity between each word in the word bank and the initial word is calculated according to the word vector of the initial word and the word vector of each word in the word bank, and the adjacent word of the initial word to be pushed to the user is determined based on the obtained similarity, so that the adjacent word of the initial word has high correlation with the initial word, and in the process of learning the word by the user, the user can conveniently memorize the word by the correlation between the words through the correlation between the words, thereby improving the efficiency of memorizing the word by the user.
The word push method provided by the present disclosure is explained in detail by specific embodiments below.
The method of the embodiment of the disclosure is applied to the intelligent terminal, can be implemented by the intelligent terminal, and in the actual use process, the intelligent terminal can be a computer, an intelligent mobile phone and the like.
Referring to fig. 1, fig. 1 is a flowchart illustrating a word pushing method according to an embodiment of the present disclosure, where the word pushing method includes the following steps S11-S14.
Step S11: a word vector for an initial word currently being pushed to a user is obtained.
The word currently pushed to the user is referred to as an initial word, and the word in the embodiment of the present disclosure may be any language vocabulary with a word structure, such as a chinese word, an english word, a japanese word, and the like, and is within the protection scope of the present application. In one example, the initial word may be the first word pushed to the user each time the user begins word learning; or the first word pushed to the user each time the user turns on the word learning software. To help the user better memorize words, the initial word may also be the last word pushed to the user the last time the user finished learning a word, or may be a word adjacent to the last word pushed to the user the last time the user finished learning a word, etc.
The word vector of the word may be a feature set, and the feature set may include one or more feature words. These feature words may be words having features that can represent the word, or the feature words themselves may have features similar to the word. The similar features may be similar word senses, similar parts of speech, similar writing methods, etc., and may be set according to actual requirements.
For example, the word "zhang san" is boy, likes eating meat, likes singing, jumping, rap, and playing basketball, the word "lie si" is girl, likes vegetables, likes singing, dancing, and swimming, the word vectors for the words "zhang san" and "lie si" may be expressed as: zhang three [ boy, eat meat, sing, dance, rap, basketball ] and Li four [ boy, vegetable, sing, dance, swimming ]. As another example, a feature word having similar features to the word "sing" may have a song, lyrics, stereo, microphone, and the word vector for the word "sing" may be expressed as: sing songs [ songs, lyrics, stereo, microphone ].
In one example, the word vector of each word may be generated by a language learning model, and the word vector of each word may be predicted by using the trained language learning model after the language learning model is trained. The Language learning Model may be an LBL Model (Log-bilingual Language Model), a CBOW Model (Continuous Bag-of-Word Model), a Skip-Gram (Skip-Word Model), a Global (global Log bilingual regression Model), or the like.
When the initial word is pushed to the user, the word vector of the initial word can be obtained according to the pushed initial word.
Step S12: and respectively calculating the similarity between each word in the word library and the initial word according to the word vector of the initial word and the word vector of each word in the word library.
Wherein the word bank includes a plurality of words and word vectors for the plurality of words.
The word vector of each word in the word bank may be preset and stored in the word bank as the attribute of the word together with the word. When the similarity between each word in the word library and the initial word needs to be calculated, the word vector of each word in the word library can be firstly obtained, and then the similarity between each word in the word library and the initial word can be calculated according to the word vector of each word in the word library and the initial word.
The similarity between each word in the word bank and the initial word can be embodied by calculating the distance between the word vector of each word in the word bank and the word vector of the initial word, and in one example, the larger the distance between the two vectors is, the more different the two words are, that is, the smaller the similarity is; a smaller distance between two vectors may indicate that the two words are more similar, i.e. the degree of similarity is greater. The distance between the word vector of each word in the word bank and the word vector of the initial word can be calculated by a distance algorithm, including but not limited to euclidean distance, manhattan distance, pearson correlation coefficient, cosine similarity, etc.
In one example, when calculating the similarity between each word in the word library and the initial word, the feature words included in the word vectors corresponding to each word and the initial word in the word library are respectively converted into numerical values to obtain a plurality of numerical value vectors corresponding to each word and the initial word in the word library, and then the similarity between each word and the initial word in the word library is calculated according to the numerical value vectors corresponding to each word and the initial word in the word library. When the characteristic words are converted into numerical values, different characteristic words can correspond to different numerical values, and the corresponding relation between the characteristic words and the numerical values can be preset.
For example, if the word vector of the word "sing" is [ song, lyric, sound, microphone ], the value 11 corresponding to "song" is preset, the value 12 corresponding to "lyric", "sound" corresponds to the value 21, and the value 22 corresponding to "microphone", the value vector of the word "sing" may be [11,12,21,22 ].
In one example, to facilitate the similarity calculation, after obtaining the numerical vectors corresponding to the words in the word bank and the initial words, the numerical vectors may be converted into binary vectors by one-hot coding (one-bit effective coding), and the distances between the words in the word bank and the binary vectors corresponding to the initial words are calculated, so as to obtain the similarities between the words in the word bank and the initial words.
In an example, the similarity may represent the similarity and the difference between the two words through a cosine value of an angle between the two vectors, in an example, a smaller angle between the two vectors may represent that the two words are more similar, that is, the similarity is larger, and conversely, a larger angle between the two vectors may represent that the two words are more different, that is, the similarity is smaller.
The cosine similarity can be calculated according to the following formula:
wherein A, B is a numerical vector of two words, Ai、BiThe numerical values in the numerical vector a and the numerical vector B, respectively.
Step S13: and determining the adjacent words corresponding to the initial words based on the similarity.
After the similarity between each word in the word library and the initial word is obtained, the word with the highest similarity to the initial word can be selected from the word library as the adjacent word corresponding to the initial word. In one example, each initial word may correspond to a neighboring word.
Step S14: and pushing the adjacent words corresponding to the initial words to the user.
After the adjacent word corresponding to the initial word is obtained, the adjacent word corresponding to the initial word may be pushed to the user when the user starts learning the next word after the user finishes learning the initial word. In one example, after recommending the neighboring words to the user, the currently recommended neighboring words may be regarded as new initial words, and the new neighboring words are determined and pushed in the manner of the above steps S11-S14.
Therefore, by applying the word pushing method provided by the disclosure, the similarity between each word in the word bank and the initial word is calculated according to the word vector of the initial word and the word vector of each word in the word bank, and the adjacent word of the initial word to be pushed to the user is determined based on the obtained similarity, so that the adjacent word of the initial word has high correlation with the initial word, and in the process of learning the word by the user, the user can conveniently memorize the word by the correlation between the words through the correlation between the words, thereby improving the efficiency of memorizing the word by the user.
In one embodiment of the present disclosure, referring to fig. 2, before the step S11 obtaining the word vector of the initial word currently pushed to the user, the method further includes:
step S21: and acquiring the multidimensional characteristics of each word in the word library.
The multidimensional feature of each word may be one or more features that can represent the characteristics of each word from different dimensions, for example, the multidimensional feature of each word may be the part of speech, the sense of speech, etc. of each word, the part of speech of each word may be the syntactic classification of each word in the language, and the sense of speech of each word may be the lexical meaning and interpretation of each word. After the multidimensional characteristics of each word can be obtained in advance, the multidimensional characteristics are stored in a word library as the attributes of the words and the words, so that the multidimensional characteristics can be directly obtained based on the word library when needed.
Step S22: and determining a word vector of each word in the word bank based on the multidimensional characteristics of each word in the word bank.
In one example, the word vector may be a feature set, and the feature set may include one or more feature words. These feature words may be words having features that can represent the word, or the feature words themselves may have features similar to the word. The multidimensional characteristic of the word can represent one or more characteristics of the word, so that characteristic words with similar characteristics to the words can be matched according to the multidimensional characteristic of each word in the word library to serve as word vectors of the words.
Step S23: word vectors for each word in the word bank are stored in the word bank.
The word vectors for each word in the word bank may be stored in the word bank as attributes of each word, along with each word and word senses of each word.
Therefore, by applying the word pushing method provided by the disclosure, the word vector of each word is determined according to the multidimensional characteristics of each word in the word library, so that the word vector of each word can more accurately represent the characteristics of the word. And the word vectors of the words are stored in the word library for subsequent application, so that the words and the word vectors corresponding to the words can be called conveniently.
In an embodiment of the present disclosure, referring to fig. 3, the step S12 above separately calculating the similarity between each word in the word library and the initial word according to the word vector of the initial word and the word vector of each word in the word library includes:
step S31: and respectively calculating the similarity between each word which is not pushed and the initial word according to the word vector of the initial word and the word vector of each word which is not pushed in the word library.
As mentioned above, the similarity between the initial word and each word in the word library can be calculated according to the word vectors corresponding to the initial word and each word in the word library, and if the similarity between each word in the word library and the initial word is directly calculated, the similarity between each word in the word library and each word in the word library can exist in the words that have been pushed
In one example, the word library may be divided into two word ranks, each word including a pushed word and an un-pushed word, and when calculating the similarity between the initial word and each word in the word library, only the similarity between each word in the word library and the initial word in the un-pushed word rank may be calculated.
In another example, each word that has been pushed in the word library may be marked as pushed to indicate that the word is a pushed word, and when calculating the similarity between the initial word and each word in the word library, only the similarity between each word in the word library that is not marked as pushed and the initial word may be calculated.
Because the word library comprises a large number of words, the word library can be used as a database, and a database index matching algorithm is adopted to realize the calculation of the similarity between the word vector of the initial word and the word vector of each un-pushed word in the word library so as to reduce the calculation amount as much as possible. For example, MySQL (relational database management system) leftmost index matching principle, etc.
After the step S14 pushes the neighboring word corresponding to the initial word to the user, the method further includes:
step S32: the neighbouring words corresponding to the initial word are marked as pushed in the word bank.
After determining the neighboring word corresponding to the initial word according to the similarity and pushing the neighboring word to the user, a label may be added to the neighboring word in the word library, the neighboring word is labeled as pushed, and the neighboring word may not be considered when the similarity is calculated next time.
As can be seen from the above, when the word pushing method provided by the present disclosure is applied and the similarity between the initial word and each word in the word library is calculated, only the similarity between each word that is not pushed in the word library and the initial word can be calculated, so as to ensure that the word pushed to the user can be a new word that is not pushed, thereby avoiding the occurrence of the situation of continuously pushing repeated words to the user, and enabling the user to perform more effective word learning.
In one possible implementation manner, referring to fig. 4, the step S13 determining the neighboring words corresponding to the initial word based on the similarities includes:
step S41: and sequencing the words which are not pushed in the word library according to the sequence of the similarity from high to low to obtain a similar word sequence corresponding to the initial word.
Step S42: and selecting the first word in the similar word sequence corresponding to the initial word to obtain the adjacent word corresponding to the initial word.
After the similarity between the initial word and each word in the word library is obtained through calculation, according to the value of each similarity, the non-pushed words corresponding to each similarity may be sorted in order from high to low to obtain a word sequence, and the word sequence may be a similar word sequence corresponding to the initial word.
Therefore, the first word in the similar word sequence corresponding to the initial word may be the word with the highest similarity to the initial word, that is, the word most similar to the initial word, and this word may be used as the adjacent word corresponding to the initial word.
Therefore, by applying the word pushing method provided by the disclosure, after similarity between each word which is not pushed and the initial word is obtained by sequencing according to a sequence from high to low, the first word in the sequence is selected as a word adjacent to the initial word, and the selected word adjacent to the initial word can be the word most similar to the initial word, so that the word pushing method can better help a user to learn words from top to bottom, and the word memorizing efficiency of the user is improved.
In an embodiment of the present disclosure, referring to fig. 5, the method further includes:
step S51: and acquiring the words which are identified by the user incorrectly or marked words to obtain key words.
In the process of learning words by a user, there may be a process of recognizing words in an attempt to help the user better memorize words. The recognized word may be a meaning of the recognized word, a part of speech of the word, a writing method of the word, a pronunciation, or the like. In the process of recognizing the word, the user may have a word recognized incorrectly. The recognition error may be a word sense, a part of speech, a writing method, a pronunciation, etc. of a misrecognized word. These words that are incorrectly recognized by the user can be recorded as key words in the user's learning process.
On the other hand, in the process of learning words by the user, there may be words that the user actively marks, and these words marked by the user are also recorded as important words in the process of learning by the user.
Step S52: and determining the scene sentence corresponding to the key word by using a preset deep learning model.
When a word is pushed to a user, scenario sentences related to the word can be pushed at the same time to help the user memorize the word, and the scenario sentences of each word can be sentences containing the word, scenario dialogs and the like. The above scenario sentences related to the words can be obtained in advance from public materials, documents, and the like on the internet.
After the key words in the learning process of the user are obtained, the scene sentence corresponding to each key word can be determined again by using the preset deep learning model.
Step S53: when the key words are pushed to the user, the scenario sentences corresponding to the key words are pushed.
When each key word is pushed to the user next time, the scenario sentence corresponding to the word and determined again by using the preset deep learning model can be pushed at the same time.
Therefore, by applying the word pushing method provided by the disclosure, the word which is identified by the user incorrectly or marked by the user can be used as the key word, and the scenario sentence corresponding to each key word is redetermined by using the preset deep learning model to be pushed to the user, so that the user can have a deeper impression on each key word, and the efficiency of memorizing the word by the user is improved.
In an embodiment of the present disclosure, referring to fig. 6, the method further includes:
step S61: and obtaining sentences in the user corpus, and training the preset deep learning model by utilizing the sentences in the user corpus.
The corpus of the user can be corpus content such as sentences which are pre-imported by the user and accord with the user's language habits, or corpus content such as sentences which are collected based on a background and are used as sentences which accord with the user's language habits in daily life.
The preset deep learning model is trained in advance by using sentences in the user corpus, so that the preset deep learning model is more consistent with the daily expression habit of the user.
Therefore, by applying the word pushing method provided by the disclosure, the preset deep learning model is trained by using the sentences in the user corpus, and the contextual sentences of the key words are obtained through the preset deep learning model, so that the contextual sentences of the key words can better accord with the daily habits of the user, the impression of the user on the key words is conveniently deepened, and the word memorizing efficiency of the user is improved.
Referring to fig. 7, the present disclosure further provides a schematic structural diagram of a word pushing apparatus, where the apparatus includes:
a word vector obtaining module 701, configured to obtain a word vector of an initial word currently pushed to a user;
a similarity calculation module 702, configured to calculate similarities between each word in the word bank and the initial word respectively according to the word vector of the initial word and the word vectors of each word in the word bank, where the word bank includes a plurality of words and word vectors of the plurality of words;
a neighboring word determining module 703, configured to determine, based on each similarity, a neighboring word corresponding to the initial word;
and a neighboring word pushing module 704, configured to push a neighboring word corresponding to the initial word to the user.
Therefore, by applying the word pushing device provided by the disclosure, the similarity between each word in the word bank and the initial word is calculated according to the word vector of the initial word and the word vector of each word in the word bank, and the adjacent word of the initial word to be pushed to the user is determined based on the obtained similarity, so that the adjacent word of the initial word has high correlation with the initial word, and in the process of learning the word by the user, the user can conveniently memorize the word by the correlation between the words through the correlation between the words, thereby improving the efficiency of memorizing the word by the user.
In an embodiment of the present disclosure, the apparatus further includes:
the semantic acquisition module is used for acquiring the multidimensional characteristics of each word in the word bank;
the word vector determining module is used for determining the word vector of each word in the word bank based on the multidimensional characteristics of each word in the word bank;
and the word vector storage module is used for storing the word vectors of all words in the word bank.
Therefore, by applying the word pushing device provided by the disclosure, the word vector of each word is determined according to the multidimensional characteristics of each word in the word library, so that the word vector of each word can more accurately represent the characteristics of the word. And the word vectors of the words are stored in the word library for subsequent application, so that the words and the word vectors corresponding to the words can be called conveniently.
In an embodiment of the present disclosure, the similarity calculation module 702 is specifically configured to:
respectively calculating the similarity of each word which is not pushed and the initial word according to the word vector of the initial word and the word vector of each word which is not pushed in the word library;
the above-mentioned device still includes:
and the word marking module is used for marking the adjacent words corresponding to the initial words in the word library as pushed words.
Therefore, when the word pushing device provided by the disclosure is applied and the similarity between the initial word and each word in the word library is calculated, the similarity between each word which is not pushed in the word library and the initial word can be only calculated, so that the word pushed to the user can be a new word which is not pushed, the situation that repeated words are continuously pushed to the user is avoided, and the user can learn more effectively.
In an embodiment of the disclosure, the neighboring word determining module 703 is specifically configured to:
sequencing each un-pushed word in the word library according to the sequence of the similarity from high to low to obtain a similar word sequence corresponding to the initial word;
and selecting the first word in the similar word sequence corresponding to the initial word to obtain the adjacent word corresponding to the initial word.
Therefore, by applying the word pushing device provided by the disclosure, after similarity between each word which is not pushed and the initial word is obtained by sequencing according to a sequence from high to low, the first word in the sequence is selected as a neighboring word of the initial word, and the selected neighboring word can be a word most similar to the initial word, so that a user can be better helped to learn words from top to bottom, and the word memorizing efficiency of the user is improved.
In an embodiment of the present disclosure, the apparatus further includes:
the key word obtaining module is used for obtaining a word which is identified by the user incorrectly or a marked word to obtain a key word;
the contextual statement determining module is used for determining contextual statements corresponding to key words by using a preset deep learning model;
and the contextual statement pushing module is used for pushing the contextual statement corresponding to the key word when the key word is pushed to the user.
Therefore, by using the word pushing device provided by the disclosure, the word which is identified by the user incorrectly or marked by the user can be used as the key word, and the situation sentence corresponding to each key word is redetermined by using the preset deep learning model to be pushed to the user, so that the user can have a deeper impression on each key word, and the efficiency of memorizing the word by the user is improved.
In an embodiment of the present disclosure, the apparatus further includes:
and the model training module is used for acquiring the sentences in the user corpus and training the preset deep learning model by using the sentences in the user corpus.
Therefore, the word pushing device provided by the disclosure is applied, the preset deep learning model is trained by the sentences in the user corpus, the contextual sentences of the key words are obtained through the preset deep learning model, the contextual sentences of the key words can be more in line with the daily habits of the user, the impression of the user on the key words is conveniently deepened, and the word memorizing efficiency of the user is improved.
In the technical scheme of the present disclosure, the processes of collecting, storing, using, processing, transmitting, providing, disclosing and the like of the personal information of the related user all conform to the regulations of the related laws and regulations, and do not violate the obtaining, storing, applying and the like of the personal information of the related user in the good custom of the public order, and all conform to the regulations of the related laws and regulations, and do not violate the good custom of the public order.
The present disclosure also provides an electronic device, a readable storage medium, and a computer program product according to embodiments of the present disclosure.
FIG. 8 illustrates a schematic block diagram of an example electronic device 800 that can be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the disclosure described and/or claimed herein.
As shown in fig. 8, the apparatus 800 includes a computing unit 801 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM)802 or a computer program loaded from a storage unit 808 into a Random Access Memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the device 800 can also be stored. The calculation unit 801, the ROM 802, and the RAM 803 are connected to each other by a bus 804. An input/output (I/O) interface 805 is also connected to bus 804.
A number of components in the device 800 are connected to the I/O interface 805, including: an input unit 806, such as a keyboard, a mouse, or the like; an output unit 807 such as various types of displays, speakers, and the like; a storage unit 808, such as a magnetic disk, optical disk, or the like; and a communication unit 809 such as a network card, modem, wireless communication transceiver, etc. The communication unit 809 allows the device 800 to exchange information/data with other devices via a computer network such as the internet and/or various telecommunication networks.
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), system on a chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.
Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions/operations specified in the flowchart and/or block diagram to be performed. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic, speech, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.
The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server with a combined blockchain.
It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present disclosure may be executed in parallel, sequentially, or in different orders, as long as the desired results of the technical solutions disclosed in the present disclosure can be achieved, and the present disclosure is not limited herein.
The above detailed description should not be construed as limiting the scope of the disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made in accordance with design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present disclosure should be included in the scope of protection of the present disclosure.
Claims (15)
1. A word push method, comprising:
acquiring a word vector of an initial word currently pushed to a user;
respectively calculating the similarity between each word in the word bank and the initial word according to the word vector of the initial word and the word vector of each word in the word bank, wherein the word bank comprises a plurality of words and word vectors of the plurality of words;
determining adjacent words corresponding to the initial words based on the similarity;
and pushing the adjacent word corresponding to the initial word to the user.
2. The method of claim 1, prior to said obtaining a word vector for an initial word currently being pushed to a user, the method further comprising:
acquiring multidimensional characteristics of each word in the word library;
determining word vectors of the words in the word bank based on the multidimensional characteristics of the words in the word bank;
and storing the word vector of each word in the word library.
3. The method of claim 1, wherein said separately calculating the similarity of each word in the word bank to the initial word based on the word vector of the initial word and the word vectors of each word in the word bank comprises:
respectively calculating the similarity of each word which is not pushed to the initial word according to the word vector of the initial word and the word vector of each word which is not pushed in the word library;
after the pushing of the neighboring word corresponding to the initial word to the user, the method further includes:
and marking the adjacent words corresponding to the initial words as pushed in the word stock.
4. The method of claim 3, wherein said determining neighboring words to which the initial word corresponds based on each of the similarities comprises:
sequencing the words which are not pushed in the word library according to the sequence of the similarity from high to low to obtain a similar word sequence corresponding to the initial word;
and selecting the first word in the similar word sequence corresponding to the initial word to obtain the adjacent word corresponding to the initial word.
5. The method of claim 1, further comprising:
acquiring a word identified by the user in error or a marked word to obtain a key word;
determining a scene sentence corresponding to the key word by using a preset deep learning model;
and when the key words are pushed to the user, pushing the scene sentences corresponding to the key words.
6. The method of claim 5, further comprising:
obtaining sentences in a user corpus, and training a preset deep learning model by utilizing the sentences in the user corpus.
7. A word pushing device comprising:
the word vector acquisition module is used for acquiring a word vector of an initial word currently pushed to a user;
the similarity calculation module is used for respectively calculating the similarity between each word in the word bank and the initial word according to the word vector of the initial word and the word vector of each word in the word bank, wherein the word bank comprises a plurality of words and word vectors of the plurality of words;
a neighboring word determining module, configured to determine, based on each of the similarities, a neighboring word corresponding to the initial word;
and the adjacent word pushing module is used for pushing the adjacent word corresponding to the initial word to the user.
8. The apparatus of claim 7, further comprising:
the characteristic acquisition module is used for acquiring multidimensional characteristics of each word in the word bank;
the word vector determining module is used for determining the word vector of each word in the word bank based on the multidimensional characteristics of each word in the word bank;
and the word vector storage module is used for storing the word vectors of all words in the word bank.
9. The apparatus of claim 7, wherein the similarity calculation module is specifically configured to:
respectively calculating the similarity of each word which is not pushed to the initial word according to the word vector of the initial word and the word vector of each word which is not pushed in the word library;
the device further comprises:
and the word marking module is used for marking the adjacent words corresponding to the initial words as pushed words in the word library.
10. The apparatus of claim 9, wherein the neighboring word determination module is specifically configured to:
sequencing the words which are not pushed in the word library according to the sequence of the similarity from high to low to obtain a similar word sequence corresponding to the initial word;
and selecting the first word in the similar word sequence corresponding to the initial word to obtain the adjacent word corresponding to the initial word.
11. The apparatus of claim 7, further comprising:
the key word obtaining module is used for obtaining a word which is identified by the user incorrectly or a marked word to obtain a key word;
the contextual statement determining module is used for determining the contextual statement corresponding to the key word by using a preset deep learning model;
and the contextual statement pushing module is used for pushing the contextual statement corresponding to the key word when the key word is pushed to the user.
12. The apparatus of claim 11, the apparatus further comprising:
and the model training module is used for acquiring the sentences in the user corpus and training the preset deep learning model by using the sentences in the user corpus.
13. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6.
14. A non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform the method of any one of claims 1-6.
15. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1-6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110995589.3A CN113705221B (en) | 2021-08-27 | 2021-08-27 | Word pushing method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110995589.3A CN113705221B (en) | 2021-08-27 | 2021-08-27 | Word pushing method and device, electronic equipment and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113705221A true CN113705221A (en) | 2021-11-26 |
CN113705221B CN113705221B (en) | 2023-11-10 |
Family
ID=78655964
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110995589.3A Active CN113705221B (en) | 2021-08-27 | 2021-08-27 | Word pushing method and device, electronic equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113705221B (en) |
Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TW539961B (en) * | 2000-12-21 | 2003-07-01 | Inventec Besta Co Ltd | Vocabulary memorizing method and system |
KR101080092B1 (en) * | 2010-11-04 | 2011-11-04 | 브레인보카주식회사 | Method for studying foreign language word and installation using thereof |
CN103116648A (en) * | 2013-02-28 | 2013-05-22 | 魏骁勇 | Word remembering method and device based on graphical context and machine learning |
CN103902527A (en) * | 2014-03-31 | 2014-07-02 | 大连大学 | Method for generating distractors of English similar word forms by being combined with word class |
CN105760356A (en) * | 2016-03-17 | 2016-07-13 | 广东小天才科技有限公司 | Automatic generation method and system of English word dictating question alternative options |
CN110047342A (en) * | 2019-04-09 | 2019-07-23 | 何基谋 | A kind of method of high efficient memory word, system and word training device |
CN110555091A (en) * | 2019-08-29 | 2019-12-10 | 韶关市启之信息技术有限公司 | Associated word generation method and device based on word vectors |
CN110717340A (en) * | 2019-09-29 | 2020-01-21 | 百度在线网络技术(北京)有限公司 | Recommendation method and device, electronic equipment and storage medium |
US20200104367A1 (en) * | 2018-09-30 | 2020-04-02 | International Business Machines Corporation | Vector Representation Based on Context |
KR102132142B1 (en) * | 2020-03-24 | 2020-07-09 | 주식회사 데이터스트림즈 | Method and apparatus for recommending vocabulary from data dictionary based on natural language processing technique |
US20200364301A1 (en) * | 2019-05-15 | 2020-11-19 | Adobe Inc. | Affect-enriched vector representation of words for use in machine-learning models |
CN112185416A (en) * | 2020-09-28 | 2021-01-05 | 上海松鼠课堂人工智能科技有限公司 | AR-based word recitation method |
CN112860846A (en) * | 2021-01-19 | 2021-05-28 | 平安科技(深圳)有限公司 | Problem vocabulary recommendation method and device, computer equipment and storage medium |
-
2021
- 2021-08-27 CN CN202110995589.3A patent/CN113705221B/en active Active
Patent Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TW539961B (en) * | 2000-12-21 | 2003-07-01 | Inventec Besta Co Ltd | Vocabulary memorizing method and system |
KR101080092B1 (en) * | 2010-11-04 | 2011-11-04 | 브레인보카주식회사 | Method for studying foreign language word and installation using thereof |
CN103116648A (en) * | 2013-02-28 | 2013-05-22 | 魏骁勇 | Word remembering method and device based on graphical context and machine learning |
CN103902527A (en) * | 2014-03-31 | 2014-07-02 | 大连大学 | Method for generating distractors of English similar word forms by being combined with word class |
CN105760356A (en) * | 2016-03-17 | 2016-07-13 | 广东小天才科技有限公司 | Automatic generation method and system of English word dictating question alternative options |
US20200104367A1 (en) * | 2018-09-30 | 2020-04-02 | International Business Machines Corporation | Vector Representation Based on Context |
CN110047342A (en) * | 2019-04-09 | 2019-07-23 | 何基谋 | A kind of method of high efficient memory word, system and word training device |
US20200364301A1 (en) * | 2019-05-15 | 2020-11-19 | Adobe Inc. | Affect-enriched vector representation of words for use in machine-learning models |
CN110555091A (en) * | 2019-08-29 | 2019-12-10 | 韶关市启之信息技术有限公司 | Associated word generation method and device based on word vectors |
CN110717340A (en) * | 2019-09-29 | 2020-01-21 | 百度在线网络技术(北京)有限公司 | Recommendation method and device, electronic equipment and storage medium |
KR102132142B1 (en) * | 2020-03-24 | 2020-07-09 | 주식회사 데이터스트림즈 | Method and apparatus for recommending vocabulary from data dictionary based on natural language processing technique |
CN112185416A (en) * | 2020-09-28 | 2021-01-05 | 上海松鼠课堂人工智能科技有限公司 | AR-based word recitation method |
CN112860846A (en) * | 2021-01-19 | 2021-05-28 | 平安科技(深圳)有限公司 | Problem vocabulary recommendation method and device, computer equipment and storage medium |
Non-Patent Citations (2)
Title |
---|
茹韶燕;邓金钉;张志威;钟洁仪;陈国雄;谭志坚;: "基于深度学习的英语单词学习系统应用研究", 电脑编程技巧与维护, no. 02, pages 135 - 136 * |
金聪;金枢炜;: "面向图像语义分类的视觉单词集成学习方法", 电子测量技术, no. 08, pages 59 - 62 * |
Also Published As
Publication number | Publication date |
---|---|
CN113705221B (en) | 2023-11-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109241524B (en) | Semantic analysis method and device, computer-readable storage medium and electronic equipment | |
CN107480143B (en) | Method and system for segmenting conversation topics based on context correlation | |
US20190073357A1 (en) | Hybrid classifier for assigning natural language processing (nlp) inputs to domains in real-time | |
WO2017127296A1 (en) | Analyzing textual data | |
CN106407113B (en) | A kind of bug localization method based on the library Stack Overflow and commit | |
CN112395385B (en) | Text generation method and device based on artificial intelligence, computer equipment and medium | |
CN113392209B (en) | Text clustering method based on artificial intelligence, related equipment and storage medium | |
CN111709243A (en) | Knowledge extraction method and device based on deep learning | |
CN111414763A (en) | Semantic disambiguation method, device, equipment and storage device for sign language calculation | |
CN107943786B (en) | Chinese named entity recognition method and system | |
CN103324621A (en) | Method and device for correcting spelling of Thai texts | |
CN114154487A (en) | Text automatic error correction method and device, electronic equipment and storage medium | |
US11947920B2 (en) | Man-machine dialogue method and system, computer device and medium | |
WO2014036827A1 (en) | Text correcting method and user equipment | |
CN111160041A (en) | Semantic understanding method and device, electronic equipment and storage medium | |
CN111090771A (en) | Song searching method and device and computer storage medium | |
CN113850080A (en) | Rhyme word recommendation method, device, equipment and storage medium | |
CN113051368A (en) | Double-tower model training method, double-tower model searching device and electronic equipment | |
CN112036186A (en) | Corpus labeling method and device, computer storage medium and electronic equipment | |
CN113919424A (en) | Training of text processing model, text processing method, device, equipment and medium | |
CN115269768A (en) | Element text processing method and device, electronic equipment and storage medium | |
CN112949293A (en) | Similar text generation method, similar text generation device and intelligent equipment | |
Gupta et al. | Songs recommendation using context-based semantic similarity between lyrics | |
CN111680146A (en) | Method and device for determining new words, electronic equipment and readable storage medium | |
CN111428487A (en) | Model training method, lyric generation method, device, electronic equipment and medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |