CN114461914B - Professional course pushing method and system based on course platform database - Google Patents

Professional course pushing method and system based on course platform database Download PDF

Info

Publication number
CN114461914B
CN114461914B CN202210127601.3A CN202210127601A CN114461914B CN 114461914 B CN114461914 B CN 114461914B CN 202210127601 A CN202210127601 A CN 202210127601A CN 114461914 B CN114461914 B CN 114461914B
Authority
CN
China
Prior art keywords
user
professional
target information
course
vocabulary
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.)
Active
Application number
CN202210127601.3A
Other languages
Chinese (zh)
Other versions
CN114461914A (en
Inventor
张昊
卞粉英
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jiangsu Ling Hu Software Technology Co ltd
Original Assignee
Jiangsu Ling Hu Software Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Jiangsu Ling Hu Software Technology Co ltd filed Critical Jiangsu Ling Hu Software Technology Co ltd
Priority to CN202210127601.3A priority Critical patent/CN114461914B/en
Publication of CN114461914A publication Critical patent/CN114461914A/en
Application granted granted Critical
Publication of CN114461914B publication Critical patent/CN114461914B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9535—Search customisation based on user profiles and personalisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2457—Query processing with adaptation to user needs
    • G06F16/24578—Query processing with adaptation to user needs using ranking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/248—Presentation of query results
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Computational Linguistics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Medical Informatics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention provides a professional course pushing method and a system based on a course platform database, wherein the method comprises the following steps: s1, acquiring user input information in a course platform database, and constructing a user target information set; s2, matching the user target information set with all complete target information sets in the database, and obtaining matching sequences of all professional courses according to the total matching similarity; s3, generating a professional course table of the corresponding sequence based on the matching sequence, and selecting behaviors of the professional course table of the corresponding sequence by a user according to a preset interaction control method; s4, replacing professional information pre-inquired by the user in the user target information set based on the selection result of the user, generating a new complete target information set, storing the new complete target information set in a database, and outputting the selection result of the user; the method and the device are used for solving the problems that the same type of professional courses in the Internet are numerous, and users cannot quickly select the professional courses suitable for the users, so that a large amount of trial and error time is needed, and time resources of the users are wasted.

Description

Professional course pushing method and system based on course platform database
Technical Field
The invention relates to the technical field of education industry, in particular to a professional course pushing method and system based on a course platform database.
Background
Along with the rapid development of the internet, a large number of professional courses are poured into the internet, in the internet era, when a user needs to improve the self capacity, the user often adopts a mode of internet plus professional courses to learn and improve the self capacity, but because the same type of professional courses in the internet are numerous, the user often cannot quickly select the professional courses suitable for the user, a large amount of trial and error time is needed, and the time resource of the user is wasted.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a professional course pushing method and a professional course pushing system based on a course platform database, which are used for solving the problems that professional courses of the same type in the Internet are various, and users often cannot quickly select the professional course suitable for the users, so that a large amount of trial and error time is needed, and time resources of the users are wasted.
A professional course pushing method based on a course platform database comprises the following steps:
s1, acquiring user input information in a course platform database, and constructing a user target information set; the set of user target information includes: the user age, the user study, the user specialization and the professional information pre-inquired by the user;
s2, matching the user target information set with all complete target information sets in the database, and obtaining matching sequences of all professional courses according to the total matching similarity; the complete set of target information includes: user age, user scholars, user specialties, and professional tutorials selected by the user;
s3, generating a professional course table of the corresponding sequence based on the matching sequence, and selecting behaviors of the professional course table of the corresponding sequence by a user according to a preset interaction control method;
and S4, replacing the professional information pre-inquired by the user in the user target information set based on the selection result of the user, generating a new complete target information set, storing the new complete target information set in the database, and outputting the selection result of the user.
As an embodiment of the present invention, S1 specifically includes:
s11, acquiring user registration information; the user registration information includes: user age, user study, user speciality;
s12, acquiring information to be queried input by a user, and extracting professional information pre-queried by the user from the information to be queried;
and S13, integrating the information in the S11 and the S12 to construct a user target information set.
As an embodiment of the present invention, S12 specifically includes:
s121, inputting information to be queried input by a user to a pre-trained professional course vocabulary prediction extraction model to obtain a segment which relates to professional course vocabularies in the information to be queried input by the user;
s122, inputting the segments related to the professional course vocabulary into a pre-trained professional information recognition model to obtain professional information pre-queried by a user in the information to be queried;
the training process of the professional course vocabulary prediction extraction model comprises the following steps:
s1211, acquiring a first input range set for extracting a professional information range, and extracting a starting point vocabulary and an end point vocabulary surrounding each professional course vocabulary in the first input range set to obtain a starting and ending vocabulary set; the first input range set comprises a plurality of pieces of pre-collected information to be inquired input by the user;
s1212, collecting and inputting the beginning and end words into a Bert model trained in advance, and obtaining the context vocabulary expression of each word in the beginning and end word set as a second input range set;
s1213, constructing an initial prediction model;
s1214, obtaining a first training set based on the first input range set, the second input range set and the professional course vocabulary corresponding to the information to be inquired input by each user in the first input range set;
s1215, training the initial prediction model by using the first training set to obtain a professional course vocabulary prediction extraction model;
the training process of the professional information recognition model comprises the following steps:
s1221, acquiring a first input vocabulary set used for triggering professional course vocabulary detection, and performing disorder processing on each professional course vocabulary in the first input vocabulary set to obtain a disorder input vocabulary set; the first input vocabulary set comprises professional course vocabularies contained in the information to be queried input by a plurality of pre-collected users;
s1222, combining the first input vocabulary set and the out-of-order input vocabulary set to obtain a second input vocabulary set;
s1223, inputting the second input vocabulary set into a pre-trained Bert model to obtain context vocabulary expression of each professional course vocabulary in the second input vocabulary set as a third input vocabulary set;
s1224, constructing an initial classification model;
s1225, obtaining a second training set based on the second input vocabulary set, the third input vocabulary set and the professional information of the user-selected course corresponding to the information to be queried input by each user in the first input vocabulary set;
and S1226, training the initial classification model by using the second training set to obtain a professional information identification model.
As an embodiment of the present invention, S2 specifically includes:
s21, taking the age of the user in the user target information set as a first age N, and taking the age of the user in the complete target information set as a second age N;
s22, taking the user scholarship in the user target information set as a first scholarship X, and taking the user scholarship in the complete target information set as a second scholarship X; the user learns the calendar and comprises: primary and following school calendars, junior middle school calendars, high school calendars, preschool calendars, doctors and the above calendars;
s23, taking the user adept specialties in the user target information set as a first specialty Z, and taking the user adept specialties in the complete target information set as a second specialty Z;
s24, using professional information pre-inquired by the user in the user target information set as professional information Y, and using a professional course selected by the user in the complete target information set as a professional course Y;
s25, respectively calculating the age matching similarity N of the user target information set and each complete target information set in S21 to S24 n Similarity X for matching study calendar x Professional matching similarity Z z Course matching similarity Y y ；
S26, corresponding age matching similarity N of each complete target information set n Similarity X for matching study calendar x Professional matching similarity Z z Course matching similarity Y y Performing weighted average calculation to obtainTo a number of total matching similarities;
s27, sequencing the total matching similarity from large to small to obtain a matching sequence, and acquiring a complete target information set corresponding to each total matching similarity in the matching sequence, wherein the corresponding professional course selected by the user is used as a course to be selected;
and S28, sequencing the courses to be selected based on the matching sequences corresponding to the complete target information sets corresponding to the courses to be selected to obtain the matching sequences of all the professional courses.
As an embodiment of the present invention, S25 specifically includes:
s251, based on the age matching similarity calculation formula, calculating the age matching similarity N between the user target information set and each complete target information set n (ii) a The age matching similarity calculation formula is as follows:
Figure BDA0003501132960000051
s252, calculating the academic matching similarity X between the user target information set and each complete target information set based on the academic matching similarity calculation formula x (ii) a The calculation formula of the similarity of the student calendar matching is as follows:
Figure BDA0003501132960000052
s253, calculating professional matching similarity Z between the user target information set and each complete target information set based on a professional matching similarity calculation formula z (ii) a The professional matching similarity calculation formula is as follows:
Figure BDA0003501132960000053
s254, calculating the course matching similarity Y between the user target information set and each complete target information set based on a course matching similarity calculation formula y (ii) a The course matching similarity calculation formula is as follows:
Figure BDA0003501132960000054
wherein S is y For all complete target information in the databaseCentralizing the number of occurrences of professional course y, S Y The number of professional courses that have been selected by the user belonging to the professional information Y in all the complete target information sets in the database.
A professional course pushing system based on a course platform database comprises:
the acquisition module is used for acquiring user input information in a course platform database;
the construction module is used for constructing a user target information set based on the user input information; the set of user target information includes: the user age, the user study, the user specialization and the professional information pre-inquired by the user;
the matching module is used for matching the user target information set with all complete target information sets in the database and obtaining matching sequences of all professional courses according to the total matching similarity; the complete set of target information includes: user age, user scholars, user specialties, and professional tutorials selected by the user;
the professional course table generating module is used for generating a professional course table of a corresponding sequence based on the matching sequence;
the interaction module is used for the user to conduct behavior selection on the professional tutorial lists of the corresponding sequences according to a preset interaction control method;
the complete target information set generation module is used for replacing professional information pre-inquired by a user in the user target information set based on a selection result of the user to generate a new complete target information set;
the storage module is used for storing the new complete target information set to a database;
and the output module is used for outputting the selection result of the user.
As an embodiment of the present invention, the obtaining module includes:
a registration information acquisition unit for acquiring user registration information; the user registration information includes: user age, user study, user speciality;
the professional information acquisition unit is used for acquiring the information to be queried input by the user and extracting the professional information pre-queried by the user in the information to be queried;
and the integration unit is used for integrating the information in the registration information acquisition unit and the professional information acquisition unit to construct a user target information set.
As an embodiment of the invention, the professional information acquisition unit executes the following operations:
inputting information to be queried input by a user to a pre-trained professional course vocabulary prediction extraction model to obtain a segment which relates to professional course vocabularies in the information to be queried input by the user;
inputting segments related to professional course vocabularies into a pre-trained professional information recognition model to obtain professional information pre-queried by a user in the information to be queried;
the training process of the professional course vocabulary prediction extraction model comprises the following steps:
acquiring a first input range set used for extracting a professional information range, and extracting a starting point vocabulary and an end point vocabulary surrounding each professional course vocabulary in the first input range set to obtain a starting and ending vocabulary set; the first input range set comprises a plurality of pieces of pre-collected information to be inquired input by the user;
the beginning and end word sets are input into a pre-trained Bert model, and context word expressions of all words in the beginning and end word sets are obtained and serve as a second input range set;
constructing an initial prediction model;
obtaining a first training set based on the first input range set, the second input range set and a professional course vocabulary corresponding to the information to be queried input by each user in the first input range set;
training an initial prediction model by using a first training set to obtain a professional course vocabulary prediction extraction model;
the training process of the professional information recognition model comprises the following steps:
acquiring a first input vocabulary set used for triggering professional course vocabulary detection, and performing disorder processing on each professional course vocabulary in the first input vocabulary set to obtain a disorder input vocabulary set; the first input vocabulary set comprises professional course vocabularies which are contained in the information to be inquired and input by a plurality of pre-collected users;
merging the first input vocabulary set and the disordered input vocabulary set to obtain a second input vocabulary set;
inputting the second input vocabulary set into a pre-trained Bert model to obtain context vocabulary expression of each professional course vocabulary in the second input vocabulary set as a third input vocabulary set;
constructing an initial classification model;
obtaining a second training set based on the second input vocabulary set, the third input vocabulary set and professional information which belongs to the user-selected course and corresponds to the information to be queried and input by each user in the first input vocabulary set;
and training the initial classification model by using the second training set to obtain a professional information identification model.
As an embodiment of the present invention, the matching module performs operations including:
taking the age of the user in the user target information set as a first age N, and taking the age of the user in the complete target information set as a second age N;
taking the user scholars in the user target information set as a first scholars X, and taking the user scholars in the complete target information set as a second scholars X; the user learns the calendar and comprises: primary and following school calendars, junior middle school calendars, high school calendars, preschool calendars, doctors and the above calendars;
taking the user adept specialty in the user target information set as a first specialty Z, and taking the user adept specialty in the complete target information set as a second specialty Z;
professional information pre-inquired by a user in the user target information set is used as professional information Y, and a professional course selected by the user in the complete target information set is used as a professional course Y;
respectively calculating the age matching similarity N of the user target information set and each complete target information set n Similarity X for matching study calendar x Professional matching similarity Z z Course matching similarity Y y ；
For each complete target messageAge matching similarity N corresponding to information set n Similarity X for matching study calendar x Professional matching similarity Z z Course matching similarity Y y Carrying out weighted average calculation to obtain a plurality of total matching similarities;
sequencing the total matching similarity from large to small to obtain a matching sequence, acquiring a complete target information set corresponding to each total matching similarity in the matching sequence, and taking the professional course selected by the corresponding user as a course to be selected;
and sequencing the courses to be selected based on the matching sequences corresponding to the complete target information sets corresponding to the courses to be selected to obtain the matching sequences of all the professional courses.
As an embodiment of the invention, the age matching similarity N of the user target information set and each complete target information set is respectively calculated n Similarity X for matching study calendar x Professional matching similarity Z z Course matching similarity Y y The method comprises the following steps:
based on an age matching similarity calculation formula, calculating the age matching similarity N between the user target information set and each complete target information set n (ii) a The age matching similarity calculation formula is as follows:
Figure BDA0003501132960000081
calculating the academic matching similarity X between the user target information set and each complete target information set based on the academic matching similarity calculation formula x (ii) a The calculation formula of the similarity of the academic calendar matching is as follows:
Figure BDA0003501132960000091
calculating professional matching similarity Z between the user target information set and each complete target information set based on a professional matching similarity calculation formula z (ii) a The professional matching similarity calculation formula is as follows:
Figure BDA0003501132960000092
based on courseMatching a similarity calculation formula, and calculating the course matching similarity Y between the user target information set and each complete target information set y (ii) a The course matching similarity calculation formula is as follows:
Figure BDA0003501132960000093
wherein S is y The number of occurrences of the professional course y is concentrated on all the complete target information in the database, S Y The number of professional courses that have been selected by the user belonging to the professional information Y in all the complete target information sets in the database.
The invention has the beneficial effects that:
the method and the device can be used for solving the problems that the same type of professional courses in the Internet are various, and users often cannot quickly select the professional courses suitable for the users, so that a large amount of trial and error time is needed, and time resources of the users are wasted.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and drawings.
The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
FIG. 1 is a flowchart of a method for pushing a professional tutorial based on a course platform database and a system thereof according to an embodiment of the present invention;
fig. 2 is a detailed flowchart of S1 in a method and a system for pushing a professional tutorial based on a course platform database according to an embodiment of the present invention;
fig. 3 is a detailed flowchart of S12 in the method and system for pushing a professional tutorial based on a course platform database according to the embodiment of the present invention;
fig. 4 is a detailed flowchart of S121 in the method and system for pushing a professional tutorial based on a course platform database according to the embodiment of the present invention;
fig. 5 is a flowchart illustrating details of S122 in the method and system for pushing a professional tutorial based on a course platform database according to the embodiment of the present invention;
fig. 6 is a detailed flowchart of S2 in the method and system for pushing a professional course based on a course platform database according to the embodiment of the present invention;
fig. 7 is a flowchart illustrating a specific process of S25 in the method and system for pushing a professional tutorial based on a course platform database according to an embodiment of the present invention;
FIG. 8 is a schematic diagram of system modules of a professional tutorial pushing method and system based on a course platform database according to an embodiment of the present invention;
fig. 9 is a schematic block diagram of an acquisition module in a professional tutorial push method and system based on a course platform database according to an embodiment of the present invention.
Detailed Description
The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.
Referring to fig. 1, an embodiment of the present invention provides a method for pushing a professional course based on a course platform database, including:
s1, acquiring user input information in a course platform database, and constructing a user target information set; the set of user target information includes: the user age, the user study, the user specialization and the professional information pre-inquired by the user;
s2, matching the user target information set with all complete target information sets in the database, and obtaining matching sequences of all professional courses according to the total matching similarity; the complete set of target information includes: user age, user scholars, user specialties, and professional tutorials selected by the user;
s3, generating a professional course table of the corresponding sequence based on the matching sequence, and selecting behaviors of the professional course table of the corresponding sequence by a user according to a preset interaction control method;
s4, replacing professional information pre-inquired by the user in the user target information set based on the selection result of the user, generating a new complete target information set, storing the new complete target information set in a database, and outputting the selection result of the user;
the working principle of the technical scheme is as follows: taking a user as an example, firstly, acquiring any user input information in a course platform database, and constructing a user target information set of the user, wherein the user target information set includes but is not limited to user age, user academic history, user specialization and professional information pre-queried by the user, and further, the more key information included in the user target information set, the higher the subsequent pushing precision; after a user target information set is constructed, matching the user target information set with all complete target information sets in a course platform database to obtain total matching similarity, and obtaining matching sequences of all professional courses according to the total matching similarity, wherein the complete target information set comprises: the method comprises the steps that a user age, a user academic calendar, a user speciality and a professional course selected by the user preferably exist a complete target information set constructed in a prediction stage when a course platform database is put into use, after a matching sequence is generated, all professional courses participating in the matching complete target information set are sequenced based on the matching sequence, a professional course table of a corresponding sequence is generated, the professional course table is preferably arranged in a webpage content arrangement mode, then the user selects the behaviors of the professional course table of the corresponding sequence according to a preset interaction control method, a selection result is obtained according to the selection of the user, the professional information pre-inquired by the user in the user target information set is replaced by the selection result, a new complete target information set is formed and stored in the course platform database, and finally the selection result of the user is output to the user, namely the professional course finally selected by the user is output to the user;
the beneficial effects of the above technical scheme are: by comparing the key information of the target user with the key information of other users and taking the professional courses selected by other users as the comparison courses, the comparison courses with higher matching degree with the target user are selected and pushed to the target user for selection, so that the problems that the same type of professional courses in the Internet are various, the user cannot quickly select the professional courses suitable for the user, a large amount of trial and error time is needed, and the time resources of the user are wasted are solved.
Referring to fig. 2, in an embodiment, S1 specifically includes:
s11, acquiring user registration information; the user registration information includes: user age, user academic calendar, user specialization;
s12, acquiring information to be queried input by a user, and extracting professional information pre-queried by the user from the information to be queried;
s13, integrating the information in the S11 and the S12 to construct a user target information set;
the working principle and the beneficial effects of the technical scheme are as follows: s1 specifically comprises the following steps: s11, acquiring user registration information; user registration information includes, but is not limited to: user age, user study, user speciality; s12, acquiring information to be queried input by a user, and extracting professional information pre-queried by the user from the information to be queried; and S13, integrating the information in the S11 and the S12, and constructing a user target information set so as to facilitate the calculation of a subsequent matching sequence.
Referring to fig. 3-5, in an embodiment, S12 specifically includes:
s121, inputting information to be queried input by a user to a pre-trained professional course vocabulary prediction extraction model to obtain a segment which relates to professional course vocabularies in the information to be queried input by the user;
s122, inputting the segments related to the professional course vocabulary into a pre-trained professional information recognition model to obtain professional information pre-queried by a user in the information to be queried;
the training process of the professional course vocabulary prediction extraction model comprises the following steps:
s1211, acquiring a first input range set for extracting the professional information range, and extracting a starting point vocabulary and an end point vocabulary surrounding each professional course vocabulary in the first input range set to obtain a starting and ending vocabulary set; the first input range set comprises a plurality of pieces of pre-collected information to be inquired input by the user;
s1212, inputting the beginning and end word sets into a pre-trained Bert model to obtain context word expressions of each word in the beginning and end word sets as a second input range set;
s1213, constructing an initial prediction model;
s1214, obtaining a first training set based on the first input range set, the second input range set and the professional course vocabulary corresponding to the information to be queried input by each user in the first input range set;
s1215, training the initial prediction model by using the first training set to obtain a professional course vocabulary prediction extraction model;
the training process of the professional information recognition model comprises the following steps:
s1221, acquiring a first input vocabulary set used for triggering professional course vocabulary detection, and performing disorder processing on each professional course vocabulary in the first input vocabulary set to obtain a disorder input vocabulary set; the first input vocabulary set comprises professional course vocabularies which are contained in the information to be inquired and input by a plurality of pre-collected users;
s1222, combining the first input vocabulary set and the out-of-order input vocabulary set to obtain a second input vocabulary set;
s1223, inputting the second input vocabulary set into a pre-trained Bert model to obtain context vocabulary expression of each professional course vocabulary in the second input vocabulary set as a third input vocabulary set;
s1224, constructing an initial classification model;
s1225, obtaining a second training set based on the second input vocabulary set, the third input vocabulary set and the professional information of the user-selected course corresponding to the information to be queried input by each user in the first input vocabulary set;
s1226, training the initial classification model by using the second training set to obtain a professional information identification model;
the working principle of the technical scheme is as follows: in practical situations, when a user inputs information to be queried, the user does not necessarily follow the professional course vocabulary with the input standard set by the system to query, but adds a series of redundant vocabularies to participate in the query, for example, [ order finding method in linear algebra ], [ how plants perform photosynthesis to obtain nutrients ], and the like, at this time, the segments related to the professional vocabularies in the information to be queried input by the user need to be extracted quickly, and the information to be queried input by the user is input to a professional course vocabulary predicting and extracting model trained in advance through S121, so that the segments related to the professional course vocabularies in the information to be queried input by the user are obtained; for example, the method comprises the following steps of extracting a linear algebra, a rank and an algorithm from a linear algebra rank algorithm, and extracting a plant, photosynthesis and nutrients from a plant photosynthesis nutrient acquisition method, wherein the training step of the professional course vocabulary prediction extraction model comprises the following steps of: s1211, acquiring a first input range set for extracting the professional information range, and extracting a starting point vocabulary and an end point vocabulary surrounding each professional course vocabulary in the first input range set to obtain a starting and ending vocabulary set; the first input range set comprises a plurality of pieces of pre-collected information to be inquired input by a user; s1212, inputting the beginning and end vocabulary sets into a pre-trained Bert (Bidirectional Encoder retrieval from transformations) model, and obtaining context vocabulary expression of each vocabulary in the beginning and end vocabulary sets as a second input range set; s1213, constructing an initial prediction model, wherein the initial prediction model is preferably a machine learning prediction model and is beneficial to improving the prediction capability; s1214, obtaining a first training set based on the first input range set, the second input range set and the professional course vocabulary corresponding to the information to be inquired input by each user in the first input range set; s1215, training the initial prediction model by using the first training set to obtain a professional course vocabulary prediction extraction model, which specifically comprises the following steps: inputting the first input range set and the second input range set into an initial prediction model, and judging whether the output content conforms to the professional course vocabulary corresponding to the information to be inquired input by each user in the first input range set until the training result reaches the expectation to obtain a professional course vocabulary prediction extraction model; in practical situations, a user can input a word considered by the user to replace the word without knowing some professional course words, for example, the word is extracted through a professional course word prediction extraction model [ the plant performs photosynthesis through green leaves ] to obtain [ plants, green leaves, photosynthesis ], [ linear algebra, ranks, law finding ] and [ plants, photosynthesis, nutrients ] obtained through the professional course word prediction extraction model, and the like, if the query is directly performed according to the word, the query result is different from the result expected by the user, at this time, what the professional course the user wants to query is needs to be quickly determined, and segments related to the professional course words are input to a professional information recognition model trained in advance through S122 to obtain professional information to be queried by the user in the information to be queried; for example, if [ nutrients ] and [ seeking method ] non-professional curriculum vocabularies are identified from [ plants, photosynthesis, nutrients ], and [ linear algebra, ranks, and seeking method ], the vocabularies actually queried by the user are [ plants, photosynthesis ], [ linear algebra, and ranks ], and the vocabularies actually queried by the user are [ plants, chloroplasts, and photosynthetic ], wherein the training step of the professional information recognition model includes: s1221, acquiring a first input vocabulary set used for triggering professional course vocabulary detection, and performing disorder processing on each professional course vocabulary in the first input vocabulary set to obtain a disorder input vocabulary set; the first input vocabulary set comprises professional course vocabularies which are contained in the information to be inquired and input by a plurality of pre-collected users; s1222, merging the first input vocabulary set and the out-of-order input vocabulary set to obtain a second input vocabulary set; for example, the second input vocabulary includes words { [ chloroplast ], [ green leaf ], [ body green leaf ], [ leaf green ], [ etc.; s1223, inputting the second input vocabulary set into a pre-trained Bert (Bidirectional Encoder retrieval from transformations) model to obtain context vocabulary expression of each professional course vocabulary in the second input vocabulary set as a third input vocabulary set; s1224, constructing an initial classification model, wherein the initial classification model is preferably a machine learning classification model and is beneficial to improving the classification capability; s1225, obtaining a second training set based on the second input vocabulary set, the third input vocabulary set and the professional information of the user-selected course corresponding to the information to be queried input by each user in the first input vocabulary set; s1226, training the initial classification model by using the second training set to obtain a professional information identification model, specifically comprising: and inputting the second input vocabulary set and the third input vocabulary set into the initial classification model, judging whether the output content conforms to the professional information of the user-selected course corresponding to the information to be inquired input by each user in the first input vocabulary set until the training result reaches the expectation, and obtaining a professional course vocabulary prediction extraction model.
The beneficial effects of the above technical scheme are: the position of the professional course vocabulary can be judged in advance through the professional course vocabulary prediction extraction model, so that the professional course vocabulary is extracted, and the extraction accuracy of the professional course vocabulary is improved; under the normal condition, the sections which are input by the user independently and relate to the professional course vocabulary do not conform to the preset professional information, and the professional information recognition model is beneficial to recognizing the sections which are input by the user and relate to the professional course vocabulary, so that the professional information carried by the sections which relate to the professional course vocabulary is obtained, and the recognition accuracy of the professional course vocabulary is improved.
Referring to fig. 6-7, in one embodiment, S2 specifically includes:
s21, taking the age of the user in the user target information set as a first age N, and taking the age of the user in the complete target information set as a second age N;
s22, taking the user scholarship in the user target information set as a first scholarship X, and taking the user scholarship in the complete target information set as a second scholarship X; the user learns the calendar and comprises: primary and following school calendars, junior middle school calendars, high school calendars, preschool calendars, doctors and the above calendars;
s23, taking the user adept specialties in the user target information set as a first specialty Z, and taking the user adept specialties in the complete target information set as a second specialty Z;
s24, using professional information pre-inquired by a user in the user target information set as professional information Y, and using a professional course selected by the user in the complete target information set as a professional course Y;
s25, respectively calculating the age matching similarity N of the user target information set and each complete target information set in S21 to S24 n Similarity X for matching study calendar x Professional matching similarity Z z Course matching similarity Y y ；
S25 specifically comprises the following steps:
s251, based on the age matching similarity calculation formula, calculating the age matching similarity N between the user target information set and each complete target information set n (ii) a The age matching similarity calculation formula is as follows:
Figure BDA0003501132960000171
s252, calculating the academic matching similarity X between the user target information set and each complete target information set based on the academic matching similarity calculation formula x (ii) a The calculation formula of the similarity of the academic calendar matching is as follows:
Figure BDA0003501132960000172
s253, calculating professional matching similarity Z between the user target information set and each complete target information set based on a professional matching similarity calculation formula z (ii) a The professional matching similarity calculation formula is as follows:
Figure BDA0003501132960000173
s254, calculating the course matching similarity Y between the user target information set and each complete target information set based on a course matching similarity calculation formula y (ii) a The course matching similarity calculation formula is as follows:
Figure BDA0003501132960000174
wherein S is y The number of occurrences of the professional course y is concentrated on all the complete target information in the database, S Y The number of professional courses which are selected by the user and belong to the professional information Y in all the complete target information sets in the database;
s26, corresponding age matching similarity N of each complete target information set n Similarity degree X for study calendar matching x Professional matching similarity Z z Course matching similarity Y y Carrying out weighted average calculation to obtain a plurality of total matching similarities;
s27, sequencing the total matching similarity from large to small to obtain a matching sequence, and acquiring a complete target information set corresponding to each total matching similarity in the matching sequence, wherein the corresponding professional course selected by the user is used as a course to be selected;
s28, sequencing the courses to be selected based on the matching sequences corresponding to the complete target information sets corresponding to the courses to be selected to obtain the matching sequences of all professional courses;
the working principle of the technical scheme is as follows: calculating matching sequences of all professional courses which accord with a current user, and firstly, determining key information of the current user and key information of other users in a course database, wherein the key information comprises age, a study, adequacy, pre-inquiry professional information or a selected professional course; the age of a user in the user target information set is taken as a first age N, the age of the user in the complete target information set is taken as a second age N, the user study in the user target information set is taken as a first study X, the user study in the complete target information set is taken as a second study X, the user excellence in the user target information set is taken as a first specialty Z, the user excellence in the user target information set is taken as a second specialty Z, the professional information pre-queried by the user in the user target information set is taken as professional information Y, and the professional course selected by the user in the complete target information set is taken as a professional course Y, preferably, the user study comprises: after selected key information is marked in classification, age matching similarity calculation formulas, history matching similarity calculation formulas, professional matching similarity calculation formulas and course matching similarity calculation formulas are used for calculating the age matching similarity N of the user target information set and each complete target information set respectively n Similarity X for matching study calendar x Professional matching similarity Z z Course matching similarity Y y Wherein S is Y The number of the professional courses selected by the user belonging to the professional information Y in all the complete target information sets in the database, i.e. the number of the professional courses belonging to the professional information Y in all the complete target information sets in the course database, for example, if the professional information pre-queried by the current user is linear algebra, S is Y The number of professional courses belonging to linear algebra in all complete target information sets in the course database is set; s y The times of occurrence of all the complete target information in the database in the concentrated professional courses y are the times of occurrence of all the complete target information in the concentrated professional courses in the course database; then, the age matching similarity N corresponding to each complete target information set n Similarity X for matching study calendar x Professional matching similarity Z z Course matching similarity Y y Carrying out weighted average calculation to obtain a plurality of total matching similarities; the calculation method is preferably as follows:
Figure BDA0003501132960000191
where ρ is the total matching similarity, α n ，α x ，α z ，α y Respectively, an age matching weight value, a scholarly matching weight value, a professional matching weight value, a course matching weight value, and alpha n ＞＞α y ＞＞α x ≥α z ，α n +α y +α x +α z =1; after the total matching similarity corresponding to all the complete target information sets is obtained, sequencing the total matching similarity from large to small to obtain a matching sequence, and then obtaining the professional course selected by the user corresponding to the complete target information set corresponding to each total matching similarity in the matching sequence as the course to be selected; finally, sequencing the courses to be selected based on the matching sequences corresponding to the complete target information sets corresponding to the courses to be selected to obtain the matching sequences of all professional courses;
the beneficial effects of the above technical scheme are: the professional courses selected by other users with the same information as the current user in the course platform database are analyzed, and the matching sequence of the professional courses matched with the current user is calculated, so that the satisfaction degree and the adaptation degree of the current user to the professional course table provided according to the matching sequence are improved, and the time spent by the current user in searching for the adapted professional courses is greatly reduced.
Referring to fig. 8, a professional course pushing system based on a course platform database includes:
the acquisition module 1 is used for acquiring user input information in a course platform database;
the building module 2 is used for building a user target information set based on the user input information; the set of user target information includes: the user age, the user academic calendar, the user speciality and the professional information pre-queried by the user;
the matching module 3 is used for matching the user target information set with all complete target information sets in the database and obtaining matching sequences of all professional courses according to the total matching similarity; the complete set of target information includes: user age, user scholars, user specialties, and professional tutorials selected by the user;
the professional course table generating module 4 is used for generating a professional course table of a corresponding sequence based on the matching sequence;
the interaction module 5 is used for the user to conduct behavior selection on the professional tutorial lists of the corresponding sequences according to a preset interaction control method;
the complete target information set generation module 6 is used for replacing professional information pre-inquired by the user in the user target information set based on a selection result of the user to generate a new complete target information set;
the storage module 7 is used for storing the new complete target information set to a database;
and the output module 8 is used for outputting the selection result of the user.
Referring to fig. 9, in an embodiment, the obtaining module 1 includes:
a registration information acquisition unit 11 for acquiring user registration information; the user registration information includes: user age, user study, user speciality;
the professional information acquisition unit 12 is used for acquiring information to be queried input by a user and extracting professional information pre-queried by the user from the information to be queried;
and the integration unit 13 is used for integrating the information in the registration information acquisition unit and the professional information acquisition unit to construct a user target information set.
In one embodiment, the professional information obtaining unit 12 performs operations including:
inputting information to be queried input by a user to a pre-trained professional course vocabulary prediction extraction model to obtain a segment which relates to professional course vocabularies in the information to be queried input by the user;
inputting the segments related to the professional course vocabulary into a pre-trained professional information recognition model to obtain the professional information pre-queried by the user in the information to be queried;
the training process of the professional course vocabulary prediction extraction model comprises the following steps:
acquiring a first input range set used for extracting a professional information range, and extracting a starting point vocabulary and an end point vocabulary surrounding each professional course vocabulary in the first input range set to obtain a starting and ending vocabulary set; the first input range set comprises a plurality of pieces of pre-collected information to be inquired input by the user;
the beginning and end word sets are input into a pre-trained Bert model, and context word expressions of all words in the beginning and end word sets are obtained and serve as a second input range set;
constructing an initial prediction model;
obtaining a first training set based on the first input range set, the second input range set and a professional course vocabulary corresponding to the information to be queried input by each user in the first input range set;
training an initial prediction model by using a first training set to obtain a professional course vocabulary prediction extraction model;
the training process of the professional information recognition model comprises the following steps:
acquiring a first input vocabulary set used for triggering professional course vocabulary detection, and performing disorder processing on each professional course vocabulary in the first input vocabulary set to obtain a disorder input vocabulary set; the first input vocabulary set comprises professional course vocabularies which are contained in the information to be inquired and input by a plurality of pre-collected users;
combining the first input vocabulary set and the out-of-order input vocabulary set to obtain a second input vocabulary set;
inputting the second input vocabulary set into a pre-trained Bert model to obtain context vocabulary expression of each professional course vocabulary in the second input vocabulary set as a third input vocabulary set;
constructing an initial classification model;
obtaining a second training set based on the second input vocabulary set, the third input vocabulary set and professional information which belongs to the user-selected course and corresponds to the information to be queried and input by each user in the first input vocabulary set;
and training the initial classification model by using the second training set to obtain a professional information identification model.
In one embodiment, the matching module 3 performs operations comprising:
taking the age of the user in the user target information set as a first age N, and taking the age of the user in the complete target information set as a second age N;
taking the user scholars in the user target information set as a first scholars X, and taking the user scholars in the complete target information set as a second scholars X; the user learns the calendar and comprises: primary and following school calendars, junior middle school calendars, high school calendars, preschool calendars, doctors and the above calendars;
taking the user adept specialty in the user target information set as a first specialty Z, and taking the user adept specialty in the complete target information set as a second specialty Z;
professional information pre-inquired by a user in the user target information set is used as professional information Y, and a professional course selected by the user in the complete target information set is used as a professional course Y;
respectively calculating user eyesAge matching similarity N of target information set and each complete target information set n Similarity X for matching study calendar x Professional matching similarity Z z Course matching similarity Y y ；
Age matching similarity N corresponding to each complete target information set n Similarity X for matching study calendar x Professional matching similarity Z z Course matching similarity Y y Carrying out weighted average calculation to obtain a plurality of total matching similarities;
sequencing the total matching similarity from large to small to obtain a matching sequence, acquiring a complete target information set corresponding to each total matching similarity in the matching sequence, and taking the professional course selected by the corresponding user as a course to be selected;
and sequencing the courses to be selected based on the matching sequences corresponding to the complete target information sets corresponding to the courses to be selected to obtain the matching sequences of all the professional courses.
In one embodiment, the age matching similarity N of the user target information set and each complete target information set is calculated separately n Similarity X for matching study calendar x Professional matching similarity Z z Course matching similarity Y y The method comprises the following steps: based on an age matching similarity calculation formula, calculating the age matching similarity N between the user target information set and each complete target information set n (ii) a The age matching similarity calculation formula is as follows:
Figure BDA0003501132960000231
calculating the academic matching similarity X between the user target information set and each complete target information set based on the academic matching similarity calculation formula x (ii) a The calculation formula of the similarity of the student calendar matching is as follows:
Figure BDA0003501132960000232
calculating professional matching similarity Z between the user target information set and each complete target information set based on a professional matching similarity calculation formula z (ii) a The professional matching similarity calculation formula is as follows: />
Figure BDA0003501132960000233
Calculating the course matching similarity Y between the user target information set and each complete target information set based on a course matching similarity calculation formula y (ii) a The course matching similarity calculation formula is as follows: />
Figure BDA0003501132960000234
Wherein S is y The number of occurrences of the professional course y is concentrated on all the complete target information in the database, S Y The number of professional courses that have been selected by the user belonging to the professional information Y in all the complete target information sets in the database.
The working principle and the beneficial effect of different internal function modules in the professional tutorial push system based on the course platform database can refer to the working principle and the beneficial effect correspondingly mentioned in the professional tutorial push method based on the course platform database, and repeated description is not repeated here.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.

Claims (8)

1. A professional course pushing method based on a course platform database is characterized by comprising the following steps:
s1, acquiring user input information in a course platform database, and constructing a user target information set; the set of user target information includes: the user age, the user study, the user specialization and the professional information pre-inquired by the user;
s2, matching the user target information set with all complete target information sets in a database, and obtaining matching sequences of all professional courses according to total matching similarity; the complete set of target information includes: user age, user scholars, user specialties, and professional tutorials selected by the user;
the S2 specifically comprises the following steps:
s21, taking the age of the user in the user target information set as a first age N, and taking the age of the user in the complete target information set as a second age N;
s22, taking the user scholarship in the user target information set as a first scholarship X, and taking the user scholarship in the complete target information set as a second scholarship X; the user learning comprises: primary and following school calendars, junior middle school calendars, high school calendars, preschool calendars, doctors and the above calendars;
s23, taking the user specialties in the user target information set as a first speciality Z, and taking the user specialties in the complete target information set as a second speciality Z;
s24, using professional information pre-inquired by the user in the user target information set as professional information Y, and using a professional course selected by the user in the complete target information set as a professional course Y;
s25, respectively calculating the age matching similarity N of the user target information set and each complete target information set in the S21 to S24 n Similarity X for matching study calendar x Professional matching similarity Z z Course matching similarity Y y ；
S26, the age matching similarity N corresponding to each complete target information set n Similarity X for matching study calendar x Professional matching similarity Z z Course matching similarity Y y Carrying out weighted average calculation to obtain a plurality of total matching similarities;
s27, sequencing the total matching similarity from large to small to obtain a matching sequence, acquiring a complete target information set corresponding to the total matching similarity in the matching sequence, and taking the professional course selected by the corresponding user as a course to be selected;
s28, sequencing the courses to be selected based on the matching sequences corresponding to the complete target information sets corresponding to the courses to be selected to obtain the matching sequences of all professional courses;
s3, generating a professional course table of a corresponding sequence based on the matching sequence, and performing behavior selection on the professional course table of the corresponding sequence by a user according to a preset interaction control method;
and S4, replacing the professional information pre-inquired by the user in the user target information set based on the selection result of the user, generating a new complete target information set, storing the new complete target information set in the database, and outputting the selection result of the user.
2. The method as claimed in claim 1, wherein the S1 specifically includes:
s11, acquiring user registration information; the user registration information includes: user age, user study, user speciality;
s12, acquiring information to be queried input by a user, and extracting professional information pre-queried by the user from the information to be queried;
and S13, integrating the information in the S11 and the S12 to construct a user target information set.
3. The method as claimed in claim 2, wherein the S12 specifically includes:
s121, inputting the information to be queried input by the user to a pre-trained professional course vocabulary prediction extraction model to obtain a segment which relates to professional course vocabularies in the information to be queried input by the user;
s122, inputting the segments related to the professional course vocabulary into a pre-trained professional information recognition model to obtain professional information pre-queried by a user in the information to be queried;
the training process of the professional course vocabulary prediction extraction model comprises the following steps:
s1211, acquiring a first input range set for extracting a professional information range, and extracting a starting point vocabulary and an end point vocabulary surrounding each professional course vocabulary in the first input range set to obtain a beginning and end vocabulary set; the first input range set comprises a plurality of pieces of pre-collected information to be inquired input by the user;
s1212, inputting the beginning and end word sets into a pre-trained Bert model to obtain context word expressions of each word in the beginning and end word sets as a second input range set;
s1213, constructing an initial prediction model;
s1214, obtaining a first training set based on the first input range set, the second input range set and the professional course vocabulary corresponding to the information to be queried and input by each user in the first input range set;
s1215, training the initial prediction model by using the first training set to obtain a professional course vocabulary prediction extraction model;
the training process of the professional information recognition model comprises the following steps:
s1221, acquiring a first input vocabulary set used for triggering professional course vocabulary detection, and performing disorder processing on each professional course vocabulary in the first input vocabulary set to obtain a disorder input vocabulary set; the first input vocabulary set comprises a plurality of pre-collected professional course vocabularies contained in the information to be inquired input by the user;
s1222, combining the first input vocabulary set and the out-of-order input vocabulary set to obtain a second input vocabulary set;
s1223, inputting the second input vocabulary set into a pre-trained Bert model to obtain context vocabulary expression of each professional course vocabulary in the second input vocabulary set as a third input vocabulary set;
s1224, constructing an initial classification model;
s1225, obtaining a second training set based on the second input vocabulary set, the third input vocabulary set and professional information of the user-selected course corresponding to the information to be queried input by each user in the first input vocabulary set;
and S1226, training the initial classification model by using the second training set to obtain a professional information recognition model.
4. The method as claimed in claim 1, wherein the S25 specifically includes:
s251, based on an age matching similarity calculation formula, calculating the age matching similarity N of the user target information set and each complete target information set n (ii) a The age matching similarity calculation formula is as follows:
Figure FDA0003955057560000041
s252, calculating the academic matching similarity X between the user target information set and each complete target information set based on an academic matching similarity calculation formula x (ii) a The calculation formula of the degree of similarity of the student calendar matching is as follows:
Figure FDA0003955057560000042
s253, calculating professional matching similarity Z between the user target information set and each complete target information set based on a professional matching similarity calculation formula z (ii) a The professional matching similarity calculation formula is as follows:
Figure FDA0003955057560000043
s254, calculating course matching similarity Y between the user target information set and each complete target information set based on a course matching similarity calculation formula y (ii) a The course matching similarity calculation formula is as follows:
Figure FDA0003955057560000044
wherein S is y The times of occurrence of professional tutorials y in the set of all complete target information in the database, S Y The number of professional courses that have been selected by the user belonging to the professional information Y in all the complete target information sets in the database.
5. A professional course pushing system based on a course platform database is characterized by comprising:
the acquisition module is used for acquiring user input information in a course platform database;
a construction module for constructing a user target information set based on the user input information; the set of user target information includes: the user age, the user academic calendar, the user speciality and the professional information pre-queried by the user;
the matching module is used for matching the user target information set with all complete target information sets in a database and obtaining matching sequences of all professional courses according to the total matching similarity; the complete set of target information includes: user age, user scholars, user specialties, and professional tutorials selected by the user;
the matching module performs operations comprising:
taking the age of the user in the user target information set as a first age N, and taking the age of the user in the complete target information set as a second age N;
taking the user scholarly calendar in the user target information set as a first scholarly calendar X, and taking the user scholarly calendar in the complete target information set as a second scholarly calendar X; the user learning comprises: primary and following school calendars, junior middle school calendars, high school calendars, preschool calendars, doctors and the above calendars;
taking the user adept specialty in the user target information set as a first specialty Z, and taking the user adept specialty in the complete target information set as a second specialty Z;
professional information pre-inquired by a user in the user target information set is used as professional information Y, and a professional course selected by the user in the complete target information set is used as a professional course Y;
respectively calculating the age matching similarity N of the user target information set and each complete target information set n Similarity X for matching study calendar x Professional matching similarity Z z And the matching similarity Y of the courses y ；
The age matching similarity N corresponding to each complete target information set n Similarity X for matching study calendar x Professional matching similar toDegree Z z And the matching similarity Y of the courses y Carrying out weighted average calculation to obtain a plurality of total matching similarities;
sequencing the total matching similarity from large to small to obtain a matching sequence, and acquiring a complete target information set corresponding to the total matching similarity in the matching sequence, wherein the professional course selected by the corresponding user is used as a course to be selected;
sequencing the courses to be selected based on the matching sequences corresponding to the complete target information sets corresponding to the courses to be selected to obtain the matching sequences of all professional courses;
the professional course table generating module is used for generating a professional course table of a corresponding sequence based on the matching sequence;
the interaction module is used for the user to conduct behavior selection on the professional tutorial lists of the corresponding sequences according to a preset interaction control method;
the complete target information set generation module is used for replacing professional information pre-inquired by a user in the user target information set based on a selection result of the user to generate a new complete target information set;
the storage module is used for storing the new complete target information set to the database;
and the output module is used for outputting the selection result of the user.
6. The system of claim 5, wherein the obtaining module comprises:
a registration information acquisition unit for acquiring user registration information; the user registration information includes: user age, user study, user speciality;
the professional information acquisition unit is used for acquiring information to be inquired input by a user and extracting professional information pre-inquired by the user from the information to be inquired;
and the integration unit is used for integrating the information in the registration information acquisition unit and the professional information acquisition unit to construct a user target information set.
7. The system for pushing a professional tutorial based on a lesson platform database as claimed in claim 6, wherein the professional information obtaining unit performs operations including:
step one, inputting the information to be queried input by the user to a pre-trained professional course vocabulary prediction extraction model to obtain a segment which relates to professional course vocabularies in the information to be queried input by the user;
inputting the segments related to the professional course vocabulary into a pre-trained professional information recognition model to obtain professional information pre-queried by a user in the information to be queried;
the training process of the professional course vocabulary prediction extraction model comprises the following steps:
acquiring a first input range set used for extracting a professional information range, and extracting a starting point vocabulary and an end point vocabulary surrounding each professional course vocabulary in the first input range set to obtain a starting and ending vocabulary set; the first input range set comprises a plurality of pieces of pre-collected information to be queried input by a user;
inputting the beginning and end word sets into a pre-trained Bert model to obtain context word expressions of each word in the beginning and end word sets as a second input range set;
constructing an initial prediction model;
obtaining a first training set based on the first input range set, the second input range set and a professional course vocabulary corresponding to the information to be queried input by each user in the first input range set;
training the initial prediction model by using the first training set to obtain a professional course vocabulary prediction extraction model;
the training process of the professional information recognition model comprises the following steps:
acquiring a first input vocabulary set used for triggering professional course vocabulary detection, and performing disorder processing on each professional course vocabulary in the first input vocabulary set to obtain a disorder input vocabulary set; the first input vocabulary set comprises professional course vocabularies contained in the information to be queried input by a plurality of pre-collected users;
merging the first input vocabulary set and the out-of-order input vocabulary set to obtain a second input vocabulary set;
inputting the second input vocabulary set into a pre-trained Bert model to obtain context vocabulary expression of each professional course vocabulary in the second input vocabulary set as a third input vocabulary set;
constructing an initial classification model;
obtaining a second training set based on the second input vocabulary set, the third input vocabulary set and professional information of the user-selected course corresponding to the information to be queried input by each user in the first input vocabulary set;
and training the initial classification model by using the second training set to obtain a professional information identification model.
8. The system as claimed in claim 5, wherein said system calculates the similarity N of age matching between said user target information set and each complete target information set n Similarity X for matching study calendar x Professional matching similarity Z z Course matching similarity Y y The method comprises the following steps:
calculating age matching similarity N between the user target information set and each complete target information set based on an age matching similarity calculation formula n (ii) a The age matching similarity calculation formula is as follows:
Figure FDA0003955057560000081
calculating the academic matching similarity X between the user target information set and each complete target information set based on an academic matching similarity calculation formula x (ii) a The calculation formula of the degree of similarity of the student calendar matching is as follows:
Figure FDA0003955057560000082
calculating professional matching similarity Z between the user target information set and each complete target information set based on a professional matching similarity calculation formula z (ii) a The professional matching similarity calculation formula is as follows:
Figure FDA0003955057560000091
calculating the course matching similarity Y between the user target information set and each complete target information set based on a course matching similarity calculation formula y (ii) a The course matching similarity calculation formula is as follows:
Figure FDA0003955057560000092
wherein S is y The number of occurrences of the professional course y is concentrated on all the complete target information in the database, S Y The number of professional courses that have been selected by the user belonging to the professional information Y in all the complete target information sets in the database.
CN202210127601.3A 2022-02-11 2022-02-11 Professional course pushing method and system based on course platform database Active CN114461914B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210127601.3A CN114461914B (en) 2022-02-11 2022-02-11 Professional course pushing method and system based on course platform database

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210127601.3A CN114461914B (en) 2022-02-11 2022-02-11 Professional course pushing method and system based on course platform database

Publications (2)

Publication Number Publication Date
CN114461914A CN114461914A (en) 2022-05-10
CN114461914B true CN114461914B (en) 2023-03-24

Family

ID=81414013

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210127601.3A Active CN114461914B (en) 2022-02-11 2022-02-11 Professional course pushing method and system based on course platform database

Country Status (1)

Country Link
CN (1) CN114461914B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104680453A (en) * 2015-02-28 2015-06-03 北京大学 Course recommendation method and system based on students' attributes
CN106846191A (en) * 2016-11-25 2017-06-13 北京粉笔蓝天科技有限公司 A kind of method of combination of curriculums table, system and server
CN109885595A (en) * 2019-01-17 2019-06-14 平安城市建设科技(深圳)有限公司 Course recommended method, device, equipment and storage medium based on artificial intelligence
CN111079964A (en) * 2018-11-20 2020-04-28 广元量知汇科技有限公司 Online education course distribution platform based on artificial intelligence

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110610404A (en) * 2019-09-11 2019-12-24 中国联合网络通信集团有限公司 Network course recommendation method, device, system, electronic equipment and storage medium
US20210264325A1 (en) * 2020-02-21 2021-08-26 Airbnb, Inc. Education-based nomadic sequencing recommendations for satisfying a curriculum scope
CN113918806A (en) * 2021-08-30 2022-01-11 北京国电通网络技术有限公司 Method for automatically recommending training courses and related equipment
CN113961666B (en) * 2021-09-18 2022-08-23 腾讯科技（深圳）有限公司 Keyword recognition method, apparatus, device, medium, and computer program product

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104680453A (en) * 2015-02-28 2015-06-03 北京大学 Course recommendation method and system based on students' attributes
CN106846191A (en) * 2016-11-25 2017-06-13 北京粉笔蓝天科技有限公司 A kind of method of combination of curriculums table, system and server
CN111079964A (en) * 2018-11-20 2020-04-28 广元量知汇科技有限公司 Online education course distribution platform based on artificial intelligence
CN109885595A (en) * 2019-01-17 2019-06-14 平安城市建设科技(深圳)有限公司 Course recommended method, device, equipment and storage medium based on artificial intelligence

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
大数据背景下的智能课程推荐系统研究;赵泉;《信息与电脑(理论版)》;20190515(第09期);全文 *

Also Published As

Publication number Publication date
CN114461914A (en) 2022-05-10

Similar Documents

Publication Publication Date Title
CN107230174B (en) Online interactive learning system and method based on network
CN109598995B (en) Intelligent teaching system based on Bayesian knowledge tracking model
US20210125516A1 (en) Answer training device, answer training method, answer generation device, answer generation method, and program
KR101220557B1 (en) Method and system for searching mobile application using human activity knowledge database
Li et al. Neural code search evaluation dataset
CN112559723B (en) FAQ search type question-answering construction method and system based on deep learning
CN106844788B (en) Library intelligent search sorting method and system
CN111782786B (en) Multi-model fusion question-answering method, system and medium for urban brain
CN110992988B (en) Speech emotion recognition method and device based on domain confrontation
CN113342958B (en) Question-answer matching method, text matching model training method and related equipment
CN114708903A (en) Method for predicting distance between protein residues based on self-attention mechanism
CN112214688A (en) Recommendation method and device for repair course selection and server
CN111783464A (en) Electric power-oriented domain entity identification method, system and storage medium
CN114492451B (en) Text matching method, device, electronic equipment and computer readable storage medium
CN114461914B (en) Professional course pushing method and system based on course platform database
CN113642652A (en) Method, device and equipment for generating fusion model
CN110929006B (en) Data type question-answering system
CN115757464B (en) Intelligent materialized view query method based on deep reinforcement learning
CN111737402A (en) Question bank intelligent knowledge point marking system
CN117131383A (en) Method for improving search precision drainage performance of double-tower model
CN111597400A (en) Computer retrieval system and method based on way-finding algorithm
CN111444414A (en) Information retrieval model for modeling various relevant characteristics in ad-hoc retrieval task
Celikkan et al. A consolidated approach for design pattern recommendation
CN113743539B (en) Form retrieval method based on deep learning
CN113486140B (en) Knowledge question and answer matching method, device, equipment and storage 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