Agricultural online learning courseware pushing method and device, electronic equipment and storage medium

Abstract

The invention provides an agricultural online learning courseware pushing method and device, electronic equipment and a storage medium. The method comprises the steps that courseware feature keywords are extracted from all agricultural online learning courseware, and user feature keywords are extracted from historical browsing data of a user; based on a word vector model, applying the courseware feature keyword of each courseware and the user feature keyword to obtain courseware word vector representation and user word vector representation of each courseware; based on the similarity between the courseware feature keyword of each courseware and the user feature keyword, and the similarity between the courseware word vector representation of each courseware and the user word vector representation, determining the mixing similarity between each courseware and the user; and based on the mixing similarity between each courseware and the user, selecting a target courseware from each courseware and pushing the target courseware to the user. According to the method and the device, the pushing accuracy can be improved, accurate pushing can be carried out according to personalized requirements of the user, and more convenient and efficient learning experience is provided for the user.

Description

Technical Field

[0001] The present invention relates to the technical field of natural language processing, and in particular to a method, device, electronic device and storage medium for pushing agricultural online learning courseware. Background Art

[0002] With the rapid development of the Internet, online video courseware resources are becoming increasingly abundant, providing users with a wide range of learning options. However, faced with a massive amount of video courseware resources, it is often difficult for users to quickly and accurately find content that meets their learning needs and interests. Traditional video courseware push mainly relies on users' historical behavior data or simple tag matching, which has certain limitations in push accuracy and personalization. Therefore, how to push video courseware efficiently, accurately and personally has become an urgent problem to be solved. Summary of the invention

[0003] The present invention provides an agricultural online learning courseware pushing method, device, electronic equipment and storage medium, which are used to solve the defects in the prior art.

[0004] The present invention provides a method for pushing agricultural online learning courseware, comprising the following steps.

[0005] Extracting courseware feature keywords from each agricultural online learning courseware, and extracting user feature keywords from the user's historical browsing data; Based on the word vector model, the courseware feature keywords of each courseware and the user feature keywords are applied to obtain the courseware word vector representation of each courseware and the user word vector representation; the word vector model is obtained by unsupervised training based on text corpus; Determine the mixed similarity between each courseware and the user based on the similarity between the courseware feature keywords of each courseware and the user feature keywords, and the similarity between the courseware word vector representation of each courseware and the user word vector representation; Based on the mixed similarity between each courseware and the user, a target courseware is selected from each courseware and pushed to the user.

[0006] According to a method for pushing agricultural online learning courseware provided by the present invention, extracting courseware characteristic keywords from each courseware includes: Determine candidate keywords for each courseware; Based on the candidate keywords of each courseware, a directed and weighted graph of each courseware is constructed; Based on the weight of each node in the directed weighted graph of each courseware, the courseware characteristic keywords of each courseware are selected from the candidate keywords of each courseware.

[0007] According to a method for pushing agricultural online learning courseware provided by the present invention, the step of determining candidate keywords for each courseware includes: Segment the text of each courseware and tag each segmented word with part of speech; Based on the part of speech of each participle, the candidate keywords for each courseware are determined.

[0008] According to a method for pushing agricultural online learning courseware provided by the present invention, extracting user characteristic keywords from the user's historical browsing data includes: Determining candidate keywords for the historical browsing data; Based on the candidate keywords of the historical browsing data, construct a directed weighted graph of the user; The user characteristic keyword is selected from the candidate keywords of the historical browsing data based on the weight of each node in the directed weighted graph of the user, the total amount of the historical browsing data, and the amount of the historical browsing data containing the candidate keyword.

[0009] According to a method for pushing agricultural online learning courseware provided by the present invention, the method of selecting the user characteristic keyword from the candidate keywords of the historical browsing data based on the weights of each node in the directed weighted graph of the user, the total amount of the historical browsing data, and the amount of the historical browsing data containing the candidate keyword comprises: Determining a keyword inverse document frequency based on the total amount of historical browsing data and the amount of historical browsing data containing the candidate keyword; Determining initial weights of candidate keywords of the historical browsing data based on the keyword inverse document frequency and the weights of each node in the directed weighted graph of the user; Based on the browsing time corresponding to the candidate keyword of the historical browsing data, the earliest browsing time of the historical browsing data, and the latest browsing time of the historical browsing data, the initial weight of the candidate keyword of the historical browsing data is corrected to obtain the weight of the candidate keyword of the historical browsing data; The user characteristic keyword is selected from the candidate keywords of the historical browsing data based on the weights of the candidate keywords of the historical browsing data.

[0010] According to a method for pushing agricultural online learning courseware provided by the present invention, the word vector model is based on applying the courseware feature keywords of each courseware and the user feature keywords to obtain the courseware word vector representation of each courseware and the user word vector representation, including: Inputting the courseware characteristic keywords of each courseware and the user characteristic keywords into the word vector model respectively, and obtaining the word vector of the characteristic keywords of each courseware and the word vector of the user characteristic keywords; Based on the weights corresponding to the characteristic keywords of each courseware, the word vectors of the characteristic keywords of each courseware are weighted and added to obtain the courseware word vector representation of each courseware; Based on the weights corresponding to the user feature keywords, the user feature keywords are weighted and added to obtain the user word vector representation.

[0011] According to an agricultural online learning courseware push method provided by the present invention, the mixed similarity between each courseware and the user is determined based on the similarity between the courseware feature keywords of each courseware and the user feature keywords, and the similarity between the courseware word vector representation of each courseware and the user word vector representation, including: Based on preset weights, the similarities between the courseware feature keywords of each courseware and the user feature keywords, as well as the similarities between the courseware word vector representation of each courseware and the user word vector representation are weighted added to obtain the mixed similarity between each courseware and the user.

[0012] The present invention also provides an agricultural online learning courseware pushing device, comprising the following modules.

[0013] An extraction unit, used to extract courseware feature keywords from each agricultural online learning courseware, and to extract user feature keywords from the user's historical browsing data; A representation unit, used to obtain a courseware word vector representation of each courseware and a user word vector representation based on a word vector model by applying the courseware feature keywords of each courseware and the user feature keywords; the word vector model is obtained by unsupervised training based on text corpus; A determination unit, configured to determine a mixed similarity between each courseware and the user based on the similarity between the courseware characteristic keywords of each courseware and the user characteristic keywords, and the similarity between the courseware word vector representation of each courseware and the user word vector representation; The push unit is used to select a target courseware from each courseware and push it to the user based on the mixed similarity between each courseware and the user.

[0014] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for pushing agricultural online learning courseware as described above is implemented.

[0015] The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the agricultural online learning courseware pushing method as described in any one of the above is implemented.

[0016] The present invention also provides a computer program product, including a computer program, and when the computer program is executed by a processor, the method for pushing agricultural online learning courseware as described in any one of the above is implemented.

[0017] The agricultural online learning courseware push method, device, electronic device and storage medium provided by the present invention determine the mixed similarity between each courseware and the user based on the similarity between the courseware feature keywords of each courseware and the user feature keywords, and the similarity between the courseware word vector representation of each courseware and the user word vector representation, so as to realize personalized push to the user based on the mixed similarity. That is, the present invention can not only improve the accuracy of push, but also perform accurate push according to the personalized needs of the user, providing the user with a more convenient and efficient learning experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 It is a flow chart of the agricultural online learning courseware pushing method provided by the present invention.

[0020] Figure 2 It is a structural diagram of the CBOW model provided by the present invention.

[0021] Figure 3 It is a schematic diagram of the framework of the agricultural online learning courseware pushing method provided by the present invention.

[0022] Figure 4 It is a structural schematic diagram of the agricultural online learning courseware pushing device provided by the present invention.

[0023] Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] Traditional video courseware push mainly relies on users' historical behavior data or simple tag matching. This method has certain limitations in push accuracy and personalization.

[0026] In this regard, the present invention proposes a method for pushing agricultural online learning courseware, which makes full use of the structured basic information data and audio data of video courseware resources, and realizes the deep understanding and representation of video courseware resources by building a word vector model, extracting keywords, forming word vector representation and other steps. At the same time, by calculating the keyword similarity matrix between the courseware resources and the user, and the word vector similarity matrix between the courseware resources and the user, personalized push to the user is realized. The courseware push method provided by the present invention can not only improve the accuracy of push, but also accurately push according to the personalized needs of the user, providing the user with a more convenient and efficient learning experience.

[0027] Figure 1 It is a flow chart of the method for pushing agricultural online learning courseware provided by the present invention, such as Figure 1 As shown, the method includes step 110 , step 120 , step 130 and step 140 .

[0028] Step 110: extracting courseware characteristic keywords from each agricultural online learning courseware, and extracting user characteristic keywords from the user's historical browsing data.

[0029] Here, the courseware can be video courseware, audio courseware, text courseware, etc. for online agricultural learning. In the case of video courseware, the audio data in the video courseware is extracted, and the audio data is converted into text data through speech recognition. In the case of audio courseware, the audio data of the audio courseware is converted into text data through speech recognition. In the case of text-visible courseware, the electronic text data is extracted through text recognition (such as OCR recognition).

[0030] Courseware feature keywords are used to represent key information in the corresponding courseware, such as key knowledge points. Optionally, the TF-IDF value of each word in the courseware can be calculated based on TF-IDF (Term Frequency - Inverse Document Frequency). The higher the TF-IDF value, the higher the frequency of occurrence of the corresponding word in the courseware, and the greater the probability that the corresponding word is a keyword. Among them, term frequency (TF) indicates the frequency of occurrence in the courseware, and inverse document frequency (IDF) indicates the importance of the word in the corpus.

[0031] In addition, the user's historical browsing data can be understood as the information of the user's historical browsing courseware. User characteristic keywords are keywords extracted from the historical browsing data to describe or represent the user's interests, preferences or characteristics. In particular, user characteristic keywords may include courseware characteristic keywords of historical browsing courseware.

[0032] Optionally, if the user browses a certain courseware frequently, it indicates that the user has a high degree of interest in the courseware. In this case, courseware feature keywords can be extracted from the courseware as user feature keywords.

[0033] Step 120: Based on the word vector model, the courseware feature keywords and user feature keywords of each courseware are applied to obtain the courseware word vector representation and user word vector representation of each courseware; the word vector model is obtained by unsupervised training based on text corpus.

[0034] Specifically, the word vector model is a language model that learns low-dimensional word vectors containing semantic information from a large amount of text corpus in an unsupervised manner. Word vectors not only have very high processing efficiency, but the trained word vectors can also represent the semantic relationship between feature words. The text data of courseware resources is used as the corpus, and the CBOW (Continuous Bag-of-WordModel) model is used to train word vectors. The word vector dimension is initially set to 300 dimensions. Figure 2 It is a structural diagram of the CBOW model provided by the present invention, such as Figure 2 As shown in the figure, the CBOW model consists of three layers, namely the input layer, the hidden layer and the output layer. The word vector corresponding to the context-related words of a certain feature word is input into the input layer, and the word vector of the specific word is output.

[0035] In addition, the courseware word vector of each courseware represents the semantic information used to characterize the corresponding courseware feature keywords, and the user word vector represents the semantic information used to characterize the user feature keywords.

[0036] Optionally, based on the word vector model, the word vector corresponding to the courseware characteristic keyword of each courseware can be determined, and the weighted sum of each word vector can be performed to obtain the courseware word vector representation of each courseware. Based on the word vector model, the word vector corresponding to the user characteristic keyword can be determined, and the weighted sum of each word vector can be performed to obtain the user word vector representation.

[0037] Step 130: Determine the mixed similarity between each courseware and the user based on the similarity between the courseware feature keywords and the user feature keywords of each courseware, and the similarity between the courseware word vector representation and the user word vector representation of each courseware.

[0038] Specifically, the similarity between the courseware feature keywords of each courseware and the user feature keywords is used to intuitively represent the matching degree of keywords between each courseware and the user from the perspective of keywords, which can be understood as a keyword similarity matrix. The similarity between the courseware word vector representation of each courseware and the user word vector representation is used to represent the matching degree of keywords between each courseware and the user from a semantic perspective, which can be understood as a word vector similarity matrix.

[0039] Taking into account the independence of keywords, the similarity between courseware feature keywords and user feature keywords lacks semantic analysis. Therefore, the embodiment of the present invention combines the similarity between the courseware feature keywords and user feature keywords of each courseware, as well as the similarity between the courseware word vector representation and the user word vector representation of each courseware, to determine the mixed similarity between each courseware and the user, so as to more accurately characterize the matching degree between each courseware and the user.

[0040] Step 140: Based on the mixed similarity between each courseware and the user, select a target courseware from each courseware and push it to the user.

[0041] Specifically, the mixed similarity between each courseware and the user is used to characterize the matching degree between each courseware and the user, and can also be used to characterize the user's interest in each courseware. The higher the mixed similarity, the higher the user's interest in the courseware.

[0042] Optionally, the courseware may be sorted in descending order of mixed similarity, and the top k courseware may be pushed to the user.

[0043] The agricultural online learning courseware push method provided by the embodiment of the present invention determines the mixed similarity between each courseware and the user based on the similarity between the courseware feature keywords of each courseware and the user feature keywords, and the similarity between the courseware word vector representation of each courseware and the user word vector representation, so as to realize personalized push to the user based on the mixed similarity. That is, the embodiment of the present invention can not only improve the accuracy of push, but also perform accurate push according to the personalized needs of the user, providing the user with a more convenient and efficient learning experience.

[0044] Based on the above embodiment, courseware feature keywords are extracted from each courseware, including: Determine candidate keywords for each courseware; Based on the candidate keywords of each courseware, a directed and weighted graph of each courseware is constructed; Based on the weight of each node in the directed weighted graph of each courseware, the courseware characteristic keywords of each courseware are selected from the candidate keywords of each courseware.

[0045] Among them, the candidate keywords for each courseware are determined, including: Segment the text of each courseware and tag each segmented word with part of speech; Based on the part of speech of each participle, the candidate keywords for each courseware are determined.

[0046] As an optional embodiment, firstly, the courseware text T of each courseware is segmented, POS tagged, and stop words are removed to retain words with POS such as nouns, verbs, and adjectives. For example, the jieba word segmentation can be used to retain the word segmentations of n, nz, v, vd, vn, l, a, and d as candidate keywords, that is, n candidate keywords are obtained, namely .

[0047] Construct a directed weighted graph G=(V,E) for the candidate keywords of each courseware, where V represents the set of nodes, E represents the set of edges, and E is a subset of V×V. A word is selected from the words in the loop, and the co-occurrence relationship is used to construct an edge between any two points. An edge exists between two nodes only if their corresponding words co-occur in a window of length K. K represents the window size, that is, at most K words co-occur. The weight of the edge between , for a given point , is the set of points pointing to this point, For point The set of points to point to.

[0048] The weight of each node is iteratively calculated according to the following formula until convergence, and the limit value is 0.0001.

[0049]

[0050] in, The initial value can be set to 1. d is the damping coefficient, ranging from 0 to 1, representing the probability of pointing from a specific point in the graph to any other point. According to experience, the value is generally 0.85. When using the TextRank algorithm to calculate the score of each point in the graph, it is necessary to specify the initial value of the point in the graph, which can be any value, and recursively calculate until convergence.

[0051] According to the weight of each node in the directed weighted graph of each courseware, each node is sorted in reverse order to obtain the most important t words as the courseware characteristic keywords of each courseware.

[0052] On this basis, each courseware can be expressed as a courseware feature keyword :

[0053] in, It represents the jth keyword of the i-th courseware after reverse sorting. express The corresponding weight.

[0054] Based on any of the above embodiments, extracting user characteristic keywords from the user's historical browsing data includes: Determine candidate keywords from historical browsing data; Based on the candidate keywords of historical browsing data, a directed and weighted graph of users is constructed; Based on the weight of each node in the directed weighted graph of the user, the total amount of historical browsing data, and the amount of historical browsing data containing the candidate keyword, the user characteristic keyword is selected from the candidate keywords of the historical browsing data.

[0055] As an optional embodiment, the historical browsing data may include the courseware browsed by the user, and the candidate keywords of the user browsing data may be understood as the candidate keywords of the courseware browsed by the user, and the candidate keywords browsed by the user may be determined by referring to the above-mentioned method for candidate keywords of the courseware, which will not be described in detail in the embodiments of the present invention.

[0056] After determining the candidate keywords of the historical browsing data, the directed weighted graph of the user can also be constructed by referring to the method for constructing the directed weighted graph of the above courseware.

[0057] Next, based on the weights of each node in the user's directed weighted graph, the total number of historical browsing data, and the number of historical browsing data containing candidate keywords, user characteristic keywords are selected from the candidate keywords in the historical browsing data, specifically including: Determine the keyword inverse document frequency based on the total number of historical browsing data and the number of historical browsing data containing the candidate keyword; Determine the initial weights of candidate keywords of historical browsing data based on the keyword inverse document frequency and the weights of each node in the user's directed weighted graph; Based on the browsing time corresponding to the candidate keyword of the historical browsing data, the earliest browsing time of the historical browsing data, and the latest browsing time of the historical browsing data, the initial weight of the candidate keyword of the historical browsing data is corrected to obtain the weight of the candidate keyword of the historical browsing data; Based on the weights of the candidate keywords of the historical browsing data, user characteristic keywords are selected from the candidate keywords of the historical browsing data.

[0058] As an optional embodiment, after determining the candidate keywords of the historical browsing data, the following formula may be used to determine the IDF value corresponding to each candidate keyword:

[0059] in, Indicates the total number of historical browsing data of the user. Indicates that the user's historical browsing data contains candidate keywords The amount of data.

[0060] Next, the initial weight of the candidate keywords of the historical browsing data is determined based on the following formula: :

[0061] in, Indicates the first In the courseware The weight of a candidate keyword, that is, the weight of the node corresponding to the candidate keyword in the user's directed weighted graph.

[0062] As the user's historical browsing data increases, the new historical browsing data can better reflect the user's current interest characteristics than the older historical browsing data. In order to describe the user's characteristic keywords more accurately and timely, when the user's characteristic keywords are updated each time, the exponential forgetting function is used for the historical user's characteristic keywords to achieve their decay over time. The exponential forgetting function is:

[0063] in, , , They represent the latest browsing time and the earliest browsing time of the historical browsing data respectively (which can be calculated on a daily basis).

[0064] Next, based on the following formula, the initial weights of the candidate keywords of the historical browsing data are modified to obtain the weights of the candidate keywords of the historical browsing data: :

[0065] Finally, the user feature keywords of the i-th user for:

[0066] in, It represents the jth candidate keyword extracted from the historical browsing data of the i-th user, after being sorted in descending order according to the weight of the candidate keywords. Indicates candidate keywords The corresponding weight.

[0067] Based on any of the above embodiments, based on the word vector model, the courseware feature keywords and user feature keywords of each courseware are applied to obtain the courseware word vector representation and user word vector representation of each courseware, including: Input the courseware characteristic keywords and user characteristic keywords of each courseware into the word vector model respectively, and obtain the word vector of the characteristic keywords of each courseware and the word vector of the user characteristic keywords; Based on the weights corresponding to the characteristic keywords of each courseware, the word vectors of the characteristic keywords of each courseware are weighted and added to obtain the courseware word vector representation of each courseware; Based on the weights corresponding to the user feature keywords, the user feature keywords are weighted and added to obtain the user word vector representation.

[0068] As an optional embodiment, the courseware word vector of each courseware The representation can be determined based on the following formula:

[0069] in, Represents the word vector of the kth characteristic keyword of the i-th courseware, Represents the weight corresponding to the kth courseware feature keyword.

[0070] User word vector representation It can be determined based on the following formula:

[0071] in, The word vector representing the kth user feature keyword of the i-th user, Indicates the weight corresponding to the k-th user feature keyword.

[0072] Based on any of the above embodiments, based on the similarity between the courseware feature keywords of each courseware and the user feature keywords, and the similarity between the courseware word vector representation of each courseware and the user word vector representation, the mixed similarity between each courseware and the user is determined, including: Based on the preset weights, the similarities between the courseware feature keywords and the user feature keywords of each courseware, as well as the similarities between the courseware word vector representation and the user word vector representation of each courseware are weighted added to obtain the mixed similarity between each courseware and the user.

[0073] In determining the characteristic keywords of each courseware And each user feature keyword After that, you can use the courseware feature keywords of each courseware And each user feature keyword The similarity between the courseware feature keywords of each courseware and the user feature keywords is calculated. Among them, the cosine matching similarity can be used to determine the similarity between the courseware feature keywords of each courseware and the user feature keywords. :

[0074] Since the similarity between the courseware feature keywords and the user feature keywords of the above-mentioned courseware is intuitively represented by the keywords of the courseware and the user, the calculation method is simple. However, due to the independence of the keywords, the similarity between the courseware feature keywords and the user feature keywords of each courseware does not take into account the semantic level, and when the number of keywords is relatively large, the sparsity of the vector is more obvious, and the amount of calculation is also significantly increased. Since the word vector has the advantage of representing the semantic information of the keyword, the distribution of semantically similar words in the vector space is also similar. Therefore, the embodiment of the present invention combines the similarity between the courseware word vector representation of each courseware and the user word vector representation to push the courseware, which can improve the semantic analysis performance and push effect of the push algorithm.

[0075] Among them, the similarity between the courseware word vector representation of each courseware and the user word vector representation It can be calculated based on the following formula:

[0076] On this basis, based on the preset weights, the similarity between the courseware feature keywords and the user feature keywords of each courseware, as well as the similarity between the courseware word vector representation and the user word vector representation of each courseware are weighted and added to obtain the mixed similarity between each courseware and the user :

[0077] Among them, parameter a and parameter b represent weights, and their value range is [0,1]. The initial value can be 0.5, and can be adjusted according to the actual situation in practical applications.

[0078] Based on any of the above embodiments, Figure 3 Schematic diagram of the framework of the method for pushing agricultural online learning courseware provided by the present invention. Figure 3 As shown, the basic information data of the courseware is obtained, and the audio corresponding to the courseware is converted into speech to obtain the corresponding courseware text. The characteristic keywords of the courseware basic information data and the courseware file are extracted to obtain the characteristic keywords of the courseware.

[0079] Obtain the user's historical browsing data (including user basic information data and user history records), extract feature keywords from the historical browsing data, and obtain user feature keywords.

[0080] Based on the word vector model, the courseware feature keywords and user feature keywords are applied to obtain the courseware word vector representation and the user word vector representation. Among them, the word vector model is obtained by preprocessing the corpus in the corpus and performing unsupervised training based on the preprocessed text corpus.

[0081] Next, based on the similarity between the courseware feature keywords and the user feature keywords (ie, keyword similarity), and the similarity between the courseware word vector representation and the user word vector representation (ie, word vector similarity), the mixed similarity between the courseware and the user is determined.

[0082] The courseware are sorted in descending order of the mixed similarity between the courseware and the user, and a courseware recommendation list is generated according to the sorting, so as to push the courseware to the user according to the recommendation list.

[0083] The courseware pushing device provided by the present invention is described below. The courseware pushing device described below and the courseware pushing method described above can be referenced to each other.

[0084] Based on any of the above embodiments, Figure 4 : is a structural schematic diagram of the agricultural online learning courseware push device provided by the present invention, such as Figure 4 As shown, the device comprises: The extraction unit 410 is used to extract courseware characteristic keywords from each agricultural online learning courseware, and to extract user characteristic keywords from the user's historical browsing data; The representation unit 420 is used to obtain the courseware word vector representation and the user word vector representation of each courseware by applying the courseware feature keywords and the user feature keywords of each courseware based on the word vector model; the word vector model is obtained by unsupervised training based on the text corpus; A determination unit 430, configured to determine a mixed similarity between each courseware and the user based on the similarity between the courseware feature keywords of each courseware and the user feature keywords, and the similarity between the courseware word vector representation of each courseware and the user word vector representation; The push unit 440 is used to select a target courseware from each courseware and push it to the user based on the mixed similarity between each courseware and the user.

[0085] Figure 5 is a schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 5As shown, the electronic device may include: a processor (processor) 510, a communication interface (Communications Interface) 520, a memory (memory) 530 and a communication bus 540, wherein the processor 510, the communication interface 520, and the memory 530 communicate with each other through the communication bus 540. The processor 510 can call the logic instructions in the memory 530 to execute the agricultural online learning courseware pushing method, which includes: extracting courseware feature keywords from each agricultural online learning courseware, and extracting user feature keywords from the user's historical browsing data; based on the word vector model, applying the courseware feature keywords of each courseware and the user feature keywords to obtain the courseware word vector representation and user word vector representation of each courseware; the word vector model is obtained by unsupervised training based on text corpus; based on the similarity between the courseware feature keywords of each courseware and the user feature keywords, and the similarity between the courseware word vector representation of each courseware and the user word vector representation, determine the mixed similarity between each courseware and the user; based on the mixed similarity between each courseware and the user, select the target courseware from each courseware and push it to the user.

[0086] In addition, the logic instructions in the above-mentioned memory 530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0087] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the agricultural online learning courseware push method provided by the above methods, which includes: extracting courseware feature keywords from each agricultural online learning courseware, and extracting user feature keywords from the user's historical browsing data; based on a word vector model, applying the courseware feature keywords of each courseware and the user feature keywords to obtain the courseware word vector representation and user word vector representation of each courseware; the word vector model is obtained by unsupervised training based on text corpus; based on the similarity between the courseware feature keywords of each courseware and the user feature keywords, and the similarity between the courseware word vector representation of each courseware and the user word vector representation, determine the mixed similarity between each courseware and the user; based on the mixed similarity between each courseware and the user, select the target courseware from each courseware and push it to the user.

[0088] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the agricultural online learning courseware push method provided by the above-mentioned methods, the method comprising: extracting courseware feature keywords from each agricultural online learning courseware, and extracting user feature keywords from the user's historical browsing data; based on a word vector model, applying the courseware feature keywords of each courseware and the user feature keywords to obtain a courseware word vector representation and a user word vector representation of each courseware; the word vector model is obtained by unsupervised training based on text corpus; based on the similarity between the courseware feature keywords of each courseware and the user feature keywords, and the similarity between the courseware word vector representation of each courseware and the user word vector representation, determining the mixed similarity between each courseware and the user; based on the mixed similarity between each courseware and the user, selecting a target courseware from each courseware and pushing it to the user.

[0089] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0090] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0091] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for pushing agricultural online learning courseware, characterized in that: include: Extracting courseware feature keywords from each agricultural online learning courseware, and extracting user feature keywords from the user's historical browsing data; Based on the word vector model, the courseware feature keywords of each courseware and the user feature keywords are applied to obtain the courseware word vector representation of each courseware and the user word vector representation; the word vector model is obtained by unsupervised training based on text corpus; Determine the mixed similarity between each courseware and the user based on the similarity between the courseware feature keywords of each courseware and the user feature keywords, and the similarity between the courseware word vector representation of each courseware and the user word vector representation; Based on the mixed similarity between each courseware and the user, a target courseware is selected from each courseware and pushed to the user.

2. The method for pushing agricultural online learning courseware according to claim 1, characterized in that: The step of extracting characteristic keywords of courseware from each courseware includes: Determine candidate keywords for each courseware; Based on the candidate keywords of each courseware, a directed and weighted graph of each courseware is constructed; Based on the weight of each node in the directed weighted graph of each courseware, the courseware characteristic keywords of each courseware are selected from the candidate keywords of each courseware.

3. The method for pushing agricultural online learning courseware according to claim 2, characterized in that: The step of determining candidate keywords for each courseware includes: Segment the text of each courseware and tag each segmented word with part of speech; Based on the part of speech of each participle, the candidate keywords for each courseware are determined.

4. The method for pushing agricultural online learning courseware according to any one of claims 1 to 3, characterized in that: The step of extracting user characteristic keywords from the user's historical browsing data includes: Determining candidate keywords for the historical browsing data; Based on the candidate keywords of the historical browsing data, construct a directed weighted graph of the user; The user characteristic keyword is selected from the candidate keywords of the historical browsing data based on the weight of each node in the directed weighted graph of the user, the total amount of the historical browsing data, and the amount of the historical browsing data containing the candidate keyword.

5. The method for pushing agricultural online learning courseware according to claim 4, characterized in that: The selecting the user characteristic keyword from the candidate keywords of the historical browsing data based on the weight of each node in the directed weighted graph of the user, the total amount of the historical browsing data, and the amount of the historical browsing data containing the candidate keyword comprises: Determining a keyword inverse document frequency based on the total amount of historical browsing data and the amount of historical browsing data containing the candidate keyword; Determining initial weights of candidate keywords of the historical browsing data based on the keyword inverse document frequency and the weights of each node in the directed weighted graph of the user; Based on the browsing time corresponding to the candidate keyword of the historical browsing data, the earliest browsing time of the historical browsing data, and the latest browsing time of the historical browsing data, the initial weight of the candidate keyword of the historical browsing data is corrected to obtain the weight of the candidate keyword of the historical browsing data; The user characteristic keyword is selected from the candidate keywords of the historical browsing data based on the weights of the candidate keywords of the historical browsing data.

6. The method for pushing agricultural online learning courseware according to any one of claims 1 to 3, characterized in that: The word vector model is based on applying the courseware feature keywords of each courseware and the user feature keywords to obtain the courseware word vector representation and the user word vector representation of each courseware, including: Inputting the courseware characteristic keywords of each courseware and the user characteristic keywords into the word vector model respectively, and obtaining the word vector of the characteristic keywords of each courseware and the word vector of the user characteristic keywords; Based on the weights corresponding to the characteristic keywords of each courseware, the word vectors of the characteristic keywords of each courseware are weighted and added to obtain the courseware word vector representation of each courseware; Based on the weights corresponding to the user feature keywords, the user feature keywords are weighted and added to obtain the user word vector representation.

7. The method for pushing agricultural online learning courseware according to any one of claims 1 to 3, characterized in that: The determining of the mixed similarity between each courseware and the user based on the similarity between the courseware feature keywords of each courseware and the user feature keywords, and the similarity between the courseware word vector representation of each courseware and the user word vector representation, includes: Based on preset weights, the similarities between the courseware feature keywords of each courseware and the user feature keywords, as well as the similarities between the courseware word vector representation of each courseware and the user word vector representation are weighted added to obtain the mixed similarity between each courseware and the user.

8. An agricultural online learning courseware push device, characterized in that: include: An extraction unit, used to extract courseware feature keywords from each agricultural online learning courseware, and to extract user feature keywords from the user's historical browsing data; A representation unit, used to obtain a courseware word vector representation of each courseware and a user word vector representation based on a word vector model by applying the courseware feature keywords of each courseware and the user feature keywords; the word vector model is obtained by unsupervised training based on text corpus; A determination unit, configured to determine a mixed similarity between each courseware and the user based on the similarity between the courseware characteristic keywords of each courseware and the user characteristic keywords, and the similarity between the courseware word vector representation of each courseware and the user word vector representation; The push unit is used to select a target courseware from each courseware and push it to the user based on the mixed similarity between each courseware and the user.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the agricultural online learning courseware pushing method as described in any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for pushing agricultural online learning courseware as described in any one of claims 1 to 7 is implemented.

Images