A method and device for predicting demand for second-classroom activities in colleges and universities

Abstract

The present invention relates to a method and device for predicting the demand for second-classroom activities in colleges and universities, the method comprising: analyzing relevant information about second-classroom activities in a given area in each time period of a reference period to determine a time series of second-classroom activity indexes; using the time series of second-classroom activity indexes and student network behavior information in a given area in each time period of the reference period as second-classroom activity features in a given area in a time period to be predicted; based on the second-classroom activity features and a second-classroom activity demand prediction model, determining the demand for second-classroom activities in colleges and universities in a given area in a time period to be predicted. The present invention extracts a time series of second-classroom activity indexes that characterize the development of second-classroom activities in each time period of the reference period from relevant information about second-classroom activities; with the help of the fusion information of the time series of second-classroom activity indexes and student behavior big data, insight is gained into the demand situation for second-classroom activities among college students in a time period to be predicted, and a scientific and accurate prediction of the demand for second-classroom activities is achieved.

Description

Technical Field

[0001] The present invention relates to the field of second-classroom education and big data technology, and in particular to a method and device for predicting the demand for second-classroom activities in colleges and universities. Background Art

[0002] "Second Classroom" is to guide and organize students to carry out various meaningful and healthy extracurricular activities outside the teaching plan, such as political, academic, knowledge, health, entertainment, public welfare activities, etc. It can be seen that "Second Classroom" is related to the quality of higher education, quality education reform and talent training in colleges and universities. Therefore, how to scientifically predict the demand for second classroom activities and grasp the development trend of "Second Classroom" activities is particularly important.

[0003] At present, the demand analysis of second-classroom activities in colleges and universities is mainly based on student questionnaires and statistical analysis, which is easily influenced by personal subjective understanding. The forecast of the demand for second-classroom activities is not accurate, comprehensive and systematic enough. Summary of the invention

[0004] The purpose of the present invention is to provide a method and device for predicting the demand for second-classroom activities in colleges and universities, so as to solve the problem that the demand prediction for second-classroom activities is not scientific and accurate enough, thereby improving the rationality of the arrangement of second-classroom activities.

[0005] In a first aspect, an embodiment of the present invention provides a method for predicting demand for second-classroom activities in colleges and universities, comprising:

[0006] Analyze the information related to the second classroom activities in a given area in each period of the reference period to determine the time series of the second classroom activity index;

[0007] The time series of the second classroom activity index and the student network behavior information of the given area in each period of the reference period are used as the second classroom activity features of the given area in the predicted period;

[0008] Based on the second classroom activity characteristics and the second classroom activity demand prediction model, determining the second classroom activity demand of all colleges and universities in a given area during the prediction period;

[0009] The reference period is a period consisting of a plurality of consecutive periods before the period to be predicted;

[0010] The second classroom activity demand prediction model is obtained by training the second classroom activity characteristics of a given area in a historical period and the actual arrangement of second classroom activities of colleges and universities in a given area in a historical period.

[0011] According to the method for predicting the demand for second classroom activities in colleges and universities provided by the present invention, the second classroom activity related information includes: college information, student personal information, second classroom activity information and tribe information;

[0012] The second classroom activity index includes: a comprehensive activity series index, a college series activity index and a tribal series activity index;

[0013] The student network behavior information includes: the duration and number of times students visit ideological and political, entertainment video, game and educational content online;

[0014] The university information includes: the name of the university, the number of students in the university, the type of university, the level of the university and the region where the university is located;

[0015] The student’s personal information includes: the student’s name, the university to which the student belongs, the number of the second-classroom activity in which the student participated, the number of growth records the student obtained, the second-classroom credits the student obtained, and the role the student played in the activity course;

[0016] The second classroom activity information includes: the second classroom activity number, activity name, activity time, activity type, the university hosting the activity, the tribe hosting the activity, the maximum number of registrants for the activity, the actual number of registrants for the activity, the actual number of sign-ins for the activity, the activity score and the activity duration;

[0017] The tribal information includes: tribal name and tribal type;

[0018] The comprehensive activity series index includes: comprehensive activity index, index of each activity type and index of each sub-region activity;

[0019] The series of university activity indexes include: activity index of each university, activity index of each university type and activity index of each university level;

[0020] The tribal series activity index includes: activity index of each tribal type.

[0021] According to the method for predicting the demand for second classroom activities in colleges and universities provided by the present invention, the analysis of the relevant information of second classroom activities in a given area in each period of the reference period to determine the time series of second classroom activity indexes includes:

[0022] For each second classroom activity index, extract the second classroom activity related information associated therewith from the second classroom activity related information of the given area in each time period of the reference period;

[0023] Based on the extracted information about the second classroom activities, determine the weight of each evaluation indicator and the time series of the score of each evaluation indicator;

[0024] The weight of each evaluation index and the time series of each evaluation index score are processed in a weighted average manner to obtain the corresponding second classroom activity index time series.

[0025] According to the method for predicting the demand for second classroom activities in colleges and universities provided by the present invention, determining the weights of each evaluation index includes:

[0026] Quantitative calculation of each evaluation index is performed on the extracted information related to the second classroom activities corresponding to the reference period to obtain the value of each evaluation index;

[0027] The obtained values ​​of each evaluation index are used as sample data of the corresponding second classroom activity index;

[0028] Using the sample data of the corresponding second classroom activity index and the sample data of the second classroom activity index similar to the corresponding second classroom activity index, a sample set is constructed;

[0029] The entropy weight method is used to perform weight analysis on the sample set to obtain the weight of each evaluation index.

[0030] According to the method for predicting the demand for second classroom activities in colleges and universities provided by the present invention, the method for identifying the second classroom activity index similar to the corresponding second classroom activity index includes:

[0031] If the second classroom activity index to be identified and the corresponding second classroom activity index are both college type activity indexes, college level activity indexes, tribe type activity indexes, activity type indexes or district activity indexes, then the second classroom activity index to be identified is similar to the corresponding second classroom activity index.

[0032] According to the method for predicting the demand for second classroom activities in colleges and universities provided by the present invention, the step of determining the time series of the scores of each evaluation index includes:

[0033] Quantitative calculation of each evaluation index is performed on the extracted information related to the second classroom activities corresponding to each period of the reference period, and the values ​​of each evaluation index corresponding to each period of the reference period are obtained;

[0034] Based on the values ​​of the evaluation indicators, determining the value distribution of each evaluation indicator;

[0035] The values ​​of the first preset quantile, the second preset quantile and the median in the numerical distribution of each evaluation index are taken as the minimum value, the maximum value and the mean value of each evaluation index;

[0036] The values ​​of each evaluation index that are outside the minimum and maximum value intervals of each evaluation index are identified as abnormal values ​​of each evaluation index;

[0037] Replace the abnormal values ​​in the values ​​of each evaluation index corresponding to each period of the reference period with the corresponding mean value;

[0038] Based on the maximum and minimum values ​​of each evaluation index, the frontier distance method is used to score the replaced values ​​of each evaluation index corresponding to each period of the reference period;

[0039] The time series of the scores of the evaluation index values ​​corresponding to the time periods of the reference cycle are used to generate the time series of the scores of the evaluation index values.

[0040] According to the method for predicting the demand for second-classroom activities in colleges and universities provided by the present invention, the evaluation indicators include: average number of activities, average activity duration, average registration rate, average sign-in rate, average activity score, average number of activity types, average number of activities participated in per capita, average activity duration, average number of growth records per capita, average number of activity types registered per capita, proportion of people without growth records, ideological and moral level, and proportion of ideological leading type activities;

[0041] The demand for second-classroom activities in the college includes: the demand for various types of second-classroom activities held by various tribes in the college.

[0042] In a second aspect, the present invention further provides a device for predicting demand for second classroom activities in colleges and universities, the device comprising:

[0043] A first determination module is used to analyze the second classroom activity related information of a given area in each time period of the reference period to determine the second classroom activity index time series;

[0044] A definition module is used to use the time series of the second classroom activity index and the student network behavior information of the given area in each period of the reference period as the second classroom activity features of the given area in the period to be predicted;

[0045] A second determination module is used to determine the second classroom activity demand of all colleges and universities in a given area during a prediction period based on the second classroom activity characteristics and the second classroom activity demand prediction model;

[0046] The reference period is a period consisting of a plurality of consecutive periods before the period to be predicted;

[0047] The second classroom activity demand prediction model is obtained by training the second classroom activity characteristics of a given area in a historical period and the actual arrangement of second classroom activities of colleges and universities in a given area in a historical period.

[0048] In a third aspect, the present invention further 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 steps of the method for predicting the demand for second classroom activities in colleges and universities as described in the first aspect are implemented.

[0049] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for predicting the demand for second classroom activities in colleges and universities as described in the first aspect.

[0050] The method and device for predicting the demand for second classroom activities in colleges and universities provided by the present invention extract a second classroom activity index time series sequence that can macroscopically characterize the development of second classroom activities in each time period of the reference period from the second classroom activity related information in a given area in each time period of the reference period; based on the fusion information of the second classroom activity index time series sequence and the student behavior big data, the demand trend of college students for "second classroom" activities in the predicted period is gained, thereby realizing a scientific and accurate prediction of the demand for second classroom activities. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] 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.

[0052] Figure 1 It is a flow chart of the method for predicting the demand for second classroom activities in colleges and universities provided by the present invention;

[0053] Figure 2 It is a schematic diagram of the actual operation flow of the method for predicting the demand for second classroom activities in colleges and universities provided by the present invention;

[0054] Figure 3 It is a schematic diagram of the activity index curve of the second classroom in colleges and universities in the embodiment provided by the present invention;

[0055] Figure 4 It is a structural schematic diagram of the device for predicting the demand for second classroom activities in colleges and universities provided by the present invention;

[0056] Figure 5 It is a structural schematic diagram of an electronic device for realizing a method for predicting demand for second-classroom activities in colleges and universities provided by the present invention. DETAILED DESCRIPTION

[0057] 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.

[0058] Combine the following Figure 1-Figure 5 The invention describes a method and device for predicting the demand for second-classroom activities in colleges and universities.

[0059] First, as Figure 1As shown, the method for predicting the demand for second-classroom activities in colleges and universities provided by the present invention comprises:

[0060] S11, analyzing the relevant information of the second classroom activities in the given area in each period of the reference period, and determining the time series of the second classroom activity index;

[0061] It should be understood that a given region can be a broad range, such as North China, Northeast China, and South China, etc.; it can also be an administrative division, such as a province or city. Information related to second-classroom activities is a general term for information such as university information, student personal information, second-classroom activity information, and information about the tribe that holds the second-classroom, which is a specific reflection of the development of second-classroom activities. The second-classroom activity index is a general term for multiple activity indexes.

[0062] The present invention extracts features (i.e., second classroom activity index time series) that can macroscopically characterize the development of second classroom activities from the relevant information of second classroom activities in a given area in each time period of a reference period, and uses this feature to gain insight into the demand trend for "second classroom" activities among college students in a given area during a predicted time period, thereby providing assistance in realizing the prediction of "second classroom" activity demand.

[0063] S12, taking the time series of the second classroom activity index and the student network behavior information of the given area in each period of the reference period as the second classroom activity features of the given area in the period to be predicted;

[0064] In the embodiment of the present invention, the student network behavior information in a given area in each time period of the reference period is represented in the form of a time series;

[0065] Students' online behavior information can directly reflect students' preferences for various types of second classrooms and their online behavior patterns. Analyzing students' online behavior information can provide certain help in predicting the demand for "second classroom" activities.

[0066] It should be noted that the acquired information related to the second classroom activities and the students' network behavior information needs to be pre-processed by data cleaning, correction, etc., and then processed into a wide table for subsequent use.

[0067] S13, determining the second classroom activity demand of all colleges and universities in a given area during a period to be predicted based on the second classroom activity characteristics and the second classroom activity demand prediction model;

[0068] It should be noted that the demand for second-classroom activities in colleges and universities of the present invention includes: the demand for various types of second-classroom activities held by various tribes in colleges and universities. From this result, it is possible to calculate the demand for second-classroom activities in a single college, the demand for various types of activities in colleges and universities, the demand for second-classroom activities in regional colleges and universities, the demand for second-classroom activities in colleges and universities at all levels, and the demand for second-classroom activities in colleges and universities in a given region.

[0069] Of course, according to needs, it is also possible to directly use the demand for second-classroom activities in a single university, the demand for various types of activities in universities, the demand for second-classroom activities in regional universities, the demand for second-classroom activities in universities at all levels, and the demand for second-classroom activities in universities in a given region as the output of the second-classroom activity demand prediction model for direct prediction.

[0070] The reference period is a period consisting of a plurality of consecutive periods before the period to be predicted;

[0071] In the embodiment of the present invention, assuming that the time period to be predicted is time period t, the reference period is the period consisting of time period t-1 to time period tM; in application, one day is usually regarded as a time period.

[0072] The second classroom activity demand prediction model is obtained by training the second classroom activity characteristics of a given area in a historical period and the actual arrangement of second classroom activities of colleges and universities in a given area in a historical period.

[0073] In the present invention, some historical periods in which the actual development of the second classroom activities is better are selected as much as possible, and the characteristics of the second classroom activities in a given area and the actual arrangement of the second classroom activities in colleges and universities in these historical periods are used as sample data to construct a sample set;

[0074] The construction process of the second classroom activity demand prediction model is as follows:

[0075] Step 1: Divide the sample set into a training set and a test set to train and test the second classroom activity demand prediction model;

[0076] Step 2: Build the initial network model architecture, that is, build an LSTM prediction network consisting of 3 LSTM layers and a fully connected layer;

[0077] First, the model is initialized. The output h0 and cell state of LSTM prediction are initialized to 0. The hidden layer weight matrix W of LSTM prediction network is c and bias b c , the weight matrix W of the input gate i and bias b i , the matrix W of the output gate o and bias b o , the matrix W of the forget gate f and bias bf Initialized to a random number between 0 and 1.

[0078] Then, the data X of the tth time granularity is t Enter the forget gate, the forget gate is based on X t With the last output value h t-1 The memory unit state C output by the hidden layer when outputting a different network time granularity data t-1 Get the forget parameter f t :

[0079] f t =sigm(W f *[h t-1 ,X t ]+b f )

[0080] Determine the state of the memory cell C t-1 Information to be stored in and its location information t , the updated memory cell state value C t .

[0081] Location information: t =sigm(W i *[h t-1 ,X t ]+b i )

[0082] Hidden layer output value:

[0083] Updated status value:

[0084] Finally, the output gate determines the output of the hidden layer, h t The final output value is:

[0085] Output gate: O t =sigm(W o *[h t-1 ,X t ]+b o )

[0086] Output value: h t =o t *tanh(C t );

[0087] Step 3: Take the characteristics of the second-classroom activities in a given area in the historical period as the input of the LSTM prediction network, and take the actual arrangements of the second-classroom activities of all colleges and universities in the given area in the historical period as the output of the LSTM prediction network. Compare the demand forecast of the second-classroom activities of all colleges and universities in the given area in the historical period with the actual arrangements, calculate the loss value, and optimize the weights through the gradient descent algorithm to reduce the loss value, and finally obtain the trained activity demand prediction model;

[0088] Step 4: Use the test set to test the trained activity demand forecasting model, and use the mean absolute error evaluation method to adjust and optimize the weight parameters; obtain the final activity demand forecasting model.

[0089] It should be added that the model integrates the big data of colleges and universities' "second classroom transcripts" (i.e., the time series of the second classroom activity index) and the big data of student behavior, and can provide insights into the demand trends of college students for "second classroom" activities.

[0090] In addition, the actual development of the second classroom activities of all colleges and universities in a given area during the forecast period can be used as historical data and substituted into the activity demand forecasting model to optimize the model parameters and improve the forecasting effect in subsequent periods.

[0091] The method for predicting the demand for second classroom activities in colleges and universities provided by the present invention extracts a second classroom activity index time series sequence that can macroscopically characterize the development of second classroom activities in each time period of the reference period from the second classroom activity related information in a given area in each time period of the reference period; based on the fusion information of the second classroom activity index time series sequence and the student behavior big data, the demand trend of college students for "second classroom" activities in the predicted time period is gained, thereby realizing a scientific and accurate prediction of the demand for second classroom activities.

[0092] Based on the above embodiments, as an optional embodiment, the second classroom activity related information includes: university information, student personal information, second classroom activity information and tribe information;

[0093] The second classroom activity index includes: a comprehensive activity series index, a college series activity index and a tribal series activity index;

[0094] The student network behavior information includes: the duration and number of times students visit ideological and political, entertainment video, game and educational content online;

[0095] The university information includes: the name of the university, the number of students in the university, the type of university, the level of the university and the region where the university is located;

[0096] The student’s personal information includes: the student’s name, the university to which the student belongs, the number of the second-classroom activity in which the student participated, the number of growth records the student obtained, the second-classroom credits the student obtained, and the role the student played in the activity course;

[0097] The second classroom activity information includes: the second classroom activity number, activity name, activity time, activity type, the university hosting the activity, the tribe hosting the activity, the maximum number of registrants for the activity, the actual number of registrants for the activity, the actual number of sign-ins for the activity, the activity score and the activity duration;

[0098] The tribal information includes: tribal name and tribal type;

[0099] The comprehensive activity series index includes: comprehensive activity index, index of each activity type and index of each sub-region activity;

[0100] The series of university activity indexes include: activity index of each university, activity index of each university type and activity index of each university level;

[0101] The tribal series activity index includes: activity index of each tribal type.

[0102] In the present invention, a tribe is a specific group organization, for example, a class can be a tribe, and a society can also be a tribe; the tribe type is the classification of tribes, such as: society, association, student union organization, class league branch and campus league committee, etc.

[0103] Activity type index, including: ideological growth activity index, volunteer public welfare activity index, cultural and sports activity index, skill specialty activity index, social practice activity index, innovation and entrepreneurship activity index, and quality development activity index;

[0104] Sub-district activity index, including: activity index of each sub-district in a given area;

[0105] Activity index of university types, including the activity index of financial universities, the activity index of science and engineering universities, the activity index of comprehensive universities, the activity index of ethnic universities, the activity index of liberal arts universities, the activity index of agriculture and forestry universities, and the activity index of normal universities;

[0106] Activity index at the college level, including: activity index of junior colleges and undergraduate colleges;

[0107] Tribal type activity index, including: department-level tribal activity index and school-level tribal activity index;

[0108] The present invention subdivides the second classroom activity index from different dimensions, analyzes the second classroom activity-related information of a given area in each time period of a reference period in different dimensions, enriches the characteristics characterizing the development of the second classroom activities in each time period of the reference period, and improves the utilization of data; in addition, it also provides an accurate data basis for the prediction of the second classroom demand of colleges and universities in the subsequent predicted time period.

[0109] Based on the above embodiments, as an optional embodiment, the analyzing the information related to the second classroom activities in a given area in each time period of the reference period to determine the time series of the second classroom activity index includes:

[0110] For each second classroom activity index, extract the second classroom activity related information associated therewith from the second classroom activity related information of the given area in each time period of the reference period;

[0111] It can be understood that the present invention selects the corresponding data range according to the index to be calculated, and the rules are as follows:

[0112] 1. Comprehensive activity series index:

[0113] ①The range of the comprehensive activity index is: data of all colleges and universities in a given area.

[0114] ② The data range of the activity type index is: filter the data according to the activity type. For example, when calculating the thought growth activity index, only select the data with the thought growth activity type in the given area.

[0115] ③The range of data for the regional activity index is: filter the data according to the region where the university is located. For example, when calculating the activity index of a certain region, only select the university data of a certain region in a given area.

[0116] 2. University series activity index

[0117] ① The data collection range of the university activity index is: filter the data according to each university separately. For example, when calculating the activity index of University A, only select the data of University A.

[0118] ② The data range of the university type activity index is: filter the data according to each university type. For example, when calculating the activity index of comprehensive universities, only select the data of comprehensive universities in a given area.

[0119] ③The data range of the university level index is: filter the data according to the university level. For example, when calculating the undergraduate university activity index, only select the data with undergraduate university level in a given area.

[0120] 3. Tribal Series Index

[0121] The range of tribal type index is as follows: data are filtered according to tribal type, for example, when calculating the department-level tribal activity index, only data on department-level activities in a given area are selected.

[0122] Based on the extracted information about the second classroom activities, determine the weight of each evaluation indicator and the time series of the score of each evaluation indicator;

[0123] In the embodiment of the present invention, a college second classroom activity evaluation index system is built, and the index system includes three first-level indicators, namely, activity scale, activity quality, and activity structure, and there are 13 second-level indicators under the three first-level indicators; the second-level indicators here are evaluation indicators; according to the indicator description, the 13 second-level indicator values ​​corresponding to the index to be calculated in each time period are calculated respectively, and then the scores of the 13 second-level indicator values ​​corresponding to the index to be calculated in each time period are obtained by scoring. Weight is an indispensable factor in the importance of the evaluation indicator, and the weight of the second-level indicator and the score of the second-level indicator value directly determine the indicator value of the indicator to be calculated.

[0124] The weight of each evaluation index and the time series of each evaluation index score are processed in a weighted average manner to obtain the corresponding second classroom activity index time series.

[0125] It can be understood that in each period, the index values ​​in the second classroom activity index system of colleges and universities are calculated, that is, the weighted sum of the secondary indicators of the second classroom activity evaluation index system of colleges and universities corresponding to the corresponding index is expressed by the formula:

[0126]

[0127] In the above formula, T represents the index in the second classroom activity index system of colleges and universities to be calculated, W k represents the weight of the kth secondary indicator in the evaluation index system of the second classroom activities in colleges and universities, DTF k It represents the score of the kth secondary indicator in the evaluation index system of the second classroom activities in colleges and universities.

[0128] The present invention adopts the DTF algorithm to calculate the comprehensive activity series index, the college series activity index and the tribal series activity index and other college second classroom activity indexes in sequence.

[0129] The present invention constructs a systematic second-classroom activity evaluation index system, and uses the calculation results of the index system to determine the second-classroom activity index of colleges and universities. The second-classroom activity index of colleges and universities is used as the influencing feature of second-classroom activity prediction input model for training, which can comprehensively predict the demand for "second-classroom" activities in colleges and universities with higher prediction accuracy.

[0130] Based on the above embodiments, as an optional embodiment, determining the weight of each evaluation index includes:

[0131] Quantitative calculation of each evaluation index is performed on the extracted information related to the second classroom activities corresponding to the reference period to obtain the value of each evaluation index;

[0132] It can be understood that after quantitative calculation, 13 evaluation index values ​​can be obtained;

[0133] The obtained values ​​of each evaluation index are used as sample data of the corresponding second classroom activity index;

[0134] It can be understood that the sample data of the corresponding second classroom activity index is 1 row and 13 columns; each column represents an evaluation indicator.

[0135] Using the sample data of the corresponding second classroom activity index and the sample data of the second classroom activity index similar to the corresponding second classroom activity index, a sample set is constructed;

[0136] Assuming that there are N-1 second classroom activity indexes similar to the corresponding second classroom activity index, the sample set is N rows and 13 columns, with one row representing one second classroom activity index and one column representing one evaluation indicator.

[0137] The entropy weight method is used to perform weight analysis on the sample set to obtain the weight of each evaluation index.

[0138] In the technical field of the present invention, the entropy weight method is a conventional weight calculation method. The specific process of determining the indicator weight using the entropy weight method is as follows:

[0139] The first step is to calculate the information entropy of each secondary indicator data of the evaluation index system of the second classroom activities of colleges and universities, and calculate the value of each indicator after standardization (the range method or z-score method can be used) as Y ik :

[0140]

[0141]

[0142] Among them, P ik represents the weight of the kth indicator under the i-th sample, E k Represents the information entropy of the kth indicator (k=1,2…13).

[0143] The second step is to calculate the entropy weight of each indicator:

[0144]

[0145] Among them, W k Represents the weight of the kth indicator.

[0146] The present invention adopts the entropy weight method to calculate the weight of each evaluation indicator under each second classroom activity index; this weight can be set to be updated once every period of time, and does not need to be updated once for each prediction. The purpose of this setting is to reduce the amount of calculation.

[0147] Based on the above embodiments, as an optional embodiment, the method for identifying the second classroom activity index similar to the corresponding second classroom activity index includes:

[0148] If the second classroom activity index to be identified and the corresponding second classroom activity index are both college type activity indexes, college level activity indexes, tribe type activity indexes, activity type indexes or district activity indexes, then the second classroom activity index to be identified is similar to the corresponding second classroom activity index.

[0149] In the technical field of the present invention, using the activity type index to illustrate, the ideological growth activity index, volunteer public welfare activity index, cultural and sports activity index, skill specialty activity index, social practice activity index, innovation and entrepreneurship activity index and quality development activity index are all activity type indexes, and they are similar second classroom activity indexes to each other.

[0150] It is understandable that the activity comprehensive index does not have a similar second classroom activity index, and its corresponding sample set has only one sample. The present invention ensures the accuracy of weight calculation.

[0151] Based on the above embodiments, as an optional embodiment, determining the time series of the scores of each evaluation index includes:

[0152] Quantitative calculation of each evaluation index is performed on the extracted information related to the second classroom activities corresponding to each period of the reference period, and the values ​​of each evaluation index corresponding to each period of the reference period are obtained;

[0153] It can be understood that, assuming that the reference period has M time periods, then M rows and 13 columns of elements are obtained here, where one row represents one time period and one column represents one evaluation indicator;

[0154] Based on the values ​​of the evaluation indicators, determining the value distribution of each evaluation indicator;

[0155] It can be understood that by extracting in columns, 13 columns of data can be extracted; the value distribution of each column of data represents the value distribution of an evaluation index;

[0156] The values ​​of the first preset quantile, the second preset quantile and the median in the numerical distribution of each evaluation index are taken as the minimum value, the maximum value and the mean value of each evaluation index;

[0157] In the embodiment of the present invention, the first preset quantile is set to 5%, and the second preset quantile is set to 85%; then the value corresponding to the first preset quantile / second preset quantile of the kth evaluation index is defined as the worst value / best value of the evaluation index, respectively calculated as max(X k ) and min(X k ).

[0158] The values ​​of each evaluation index that are outside the minimum and maximum value intervals of each evaluation index are identified as abnormal values ​​of each evaluation index;

[0159] In order to eliminate the influence of outliers on the overall evaluation results, the embodiment of the present invention deletes the quantile values ​​less than 5% and the index values ​​corresponding to the quantile values ​​greater than 85%.

[0160] Replace the abnormal values ​​in the values ​​of each evaluation index corresponding to each period of the reference period with the corresponding mean value;

[0161] Based on the maximum and minimum values ​​of each evaluation index, the frontier distance method is used to score the replaced values ​​of each evaluation index corresponding to each period of the reference period;

[0162] The present invention scores the values ​​of each evaluation index, which can be achieved in the following way:

[0163] Positive indicators:

[0164]

[0165] Reverse Indicator:

[0166]

[0167] Among them, DTF mk represents the score of the kth evaluation index value in the mth period, X mk Represents the kth evaluation index value in the mth time period.

[0168] The distance to frontier method (DTF algorithm) is used to calculate the quantitative results of each indicator of the evaluation index system of the second classroom activities in colleges and universities. The full score is 100 points and the lowest score is 0 points. The higher the score, the better the indicator performance.

[0169] The time series of the scores of the evaluation index values ​​corresponding to the time periods of the reference cycle are used to generate the time series of the scores of the evaluation index values.

[0170] The present invention ensures the rationality of scoring, and further ensures the rationality of calculating the second classroom activity index of each university.

[0171] On the basis of the above embodiments, as an optional embodiment, the evaluation indicators include: average number of activities, average activity duration, average registration rate, average sign-in rate, average activity score, average number of activity types, average number of activities participated in per capita, average activity duration per capita, average number of growth records per capita, average number of activity types registered per capita, proportion of people with no growth record, ideological and moral level and proportion of ideological guidance type activities; the demand for second classroom activities of the university includes: the demand for various types of second classroom activities held by various tribes of the university.

[0172] In the embodiment of the present invention, the demand for second-classroom activities in colleges and universities can also be statistical demand conditions such as the demand for second-classroom activities in a single college and university, the demand for various types of activities in colleges and universities, the demand for second-classroom activities in regional colleges and universities, the demand for second-classroom activities in colleges and universities at all levels, and the demand for second-classroom activities in colleges and universities in a given region. The specific demand conditions can be selected according to needs.

[0173] In the embodiment of the present invention, the indicator description of each evaluation indicator (secondary indicator) is given in constructing the evaluation indicator system of the second classroom activities in colleges and universities, as shown in Table 1:

[0174] Table 1

[0175]

[0176]

[0177] The present invention can be carried out as follows Figure 2 The process shown is executed, namely:

[0178] S1. Obtaining the network behavior information of college students and related information of second-classroom activities: First, obtain the college information, student personal information, student network behavior information, second-classroom activity information and the tribe information of the second-classroom.

[0179] S2. Build an evaluation index system for the second classroom activities in colleges and universities, select the data range corresponding to the index to be calculated, and calculate the secondary index value according to the index description:

[0180] Among them, an evaluation index system for the second classroom activities in colleges and universities is established, which includes three first-level indicators: activity scale, activity quality, and activity structure, and 13 second-level indicators.

[0181] S3. Use the DTF algorithm to calculate the second classroom activity index of colleges and universities: calculate the comprehensive activity series index of the second classroom of colleges and universities, the college series activity index and the tribe series activity index in turn.

[0182] S4. Modeling and predicting the demand for second-classroom activities in colleges and universities: Use the time series data of college students’ network behavior, the time series data of comprehensive activity series index, the time series data of college series activity index and the time series data of tribe series activity index as model input to predict the demand for second-classroom activities.

[0183] It is assumed that the demand for second-classroom activities in a single university, the demand for various types of activities in universities, or the demand for second-classroom activities in regional universities is used as the predicted demand for second-classroom activities;

[0184] For example, the development of second-classroom activities in 100 sample universities in 2020 was measured according to the method described above. Figure 3 The time series calculation results of the comprehensive activity index, the second classroom activity index of colleges and universities in Province X, and the activity index of the second classroom activity index of University A and College B are illustrated.

[0185] In the second aspect, the device for predicting the demand for second classroom activities in colleges and universities provided by the present invention is described. The device for predicting the demand for second classroom activities in colleges and universities described below and the method for predicting the demand for second classroom activities in colleges and universities described above can be referenced to each other. Figure 4 The following is a schematic diagram of a device for predicting the demand for activities in the second classroom of a university. Figure 4 As shown, the device includes: a first determination module 21, a definition module 22 and a second determination module 23;

[0186] Among them, the first determination module 21 is used to analyze the second classroom activity related information of the given area in each time period of the reference period, and determine the second classroom activity index time series sequence; the definition module 22 is used to use the second classroom activity index time series sequence and the student network behavior information of the given area in each time period of the reference period as the second classroom activity characteristics of the given area in the predicted period; the second determination module 23 is used to determine the second classroom activity needs of all colleges and universities in the given area in the predicted period based on the second classroom activity characteristics and the second classroom activity demand prediction model; wherein, the reference period is a period composed of multiple consecutive time periods before the predicted period; the second classroom activity demand prediction model is trained by the second classroom activity characteristics of the given area in the historical period and the actual arrangement of the second classroom activities of colleges and universities in the given area in the historical period.

[0187] The device for predicting the demand for second classroom activities in colleges and universities provided by the present invention extracts a time series sequence of second classroom activity indexes that can macroscopically characterize the development of second classroom activities in each time period of the reference period from the relevant information about second classroom activities in a given area in each time period of the reference period; based on the fusion information of the time series sequence of second classroom activity indexes and the big data of student behavior, the demand trend of college students for "second classroom" activities in the time period to be predicted is gained, thereby realizing a scientific and accurate prediction of the demand for second classroom activities.

[0188] Thirdly, Figure 5 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 5 As shown, the electronic device may include: a processor 510, a communication interface 520, a 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 may call the logic instructions in the memory 530 to execute the method for predicting the demand for second classroom activities in colleges and universities, the method comprising: analyzing the relevant information about the second classroom activities in a given area in each period of the reference period to determine the time series of the second classroom activity index; using the time series of the second classroom activity index and the student network behavior information in the given area in each period of the reference period as the second classroom activity characteristics of the given area in the period to be predicted; based on the second classroom activity characteristics and the second classroom activity demand prediction model, determining the second classroom activity demand of all colleges and universities in the given area in the period to be predicted; wherein the reference period is a period composed of a plurality of consecutive periods before the period to be predicted; the second classroom activity demand prediction model is obtained by training the second classroom activity characteristics of the given area in the historical period and the actual arrangement of the second classroom activities of the colleges and universities in the given area in the historical period.

[0189] 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.

[0190] In a fourth aspect, the present invention further 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 method for predicting the demand for second classroom activities in colleges and universities provided in the above-mentioned embodiments, the method comprising: analyzing the second classroom activity-related information of a given area in each time period of a reference period, and determining a time series sequence of second classroom activity indexes; using the second classroom activity index time series sequence and the student network behavior information of a given area in each time period of the reference period as second classroom activity characteristics of the given area in the time period to be predicted; based on the second classroom activity characteristics and the second classroom activity demand prediction model, determining the demand for second classroom activities of all colleges and universities in the given area in the time period to be predicted; wherein the reference period is a period consisting of multiple consecutive time periods before the time period to be predicted; the second classroom activity demand prediction model is obtained by training the second classroom activity characteristics of a given area in a historical period and the actual arrangements for second classroom activities of colleges and universities in the given area in the historical period.

[0191] 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.

[0192] 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, it 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.

[0193] 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 predicting the demand for second-classroom activities in colleges and universities, characterized in that: The method comprises: Analyze the second classroom activity related information of the given area in each time period of the reference period to determine the second classroom activity index time series sequence, which includes: for each second classroom activity index, extract the second classroom activity related information associated with it from the second classroom activity related information of the given area in each time period of the reference period; based on the extracted second classroom activity related information, determine the weight of each evaluation indicator and the time series sequence of each evaluation indicator score; process the weight of each evaluation indicator and the time series sequence of each evaluation indicator score in a weighted average manner to obtain the corresponding second classroom activity index time series sequence; The time series of the second classroom activity index and the student network behavior information of the given area in each period of the reference period are used as the second classroom activity features of the given area in the predicted period; Based on the second classroom activity characteristics and the second classroom activity demand prediction model, determining the second classroom activity demand of all colleges and universities in a given area during the prediction period; The reference period is a period consisting of a plurality of consecutive periods before the period to be predicted; The second classroom activity demand prediction model is obtained by training the second classroom activity characteristics of a given area in a historical period and the actual arrangement of second classroom activities of colleges and universities in a given area in a historical period.

2. The method for predicting the demand for second-classroom activities in colleges and universities according to claim 1 is characterized in that: The second classroom activity related information includes: university information, student personal information, second classroom activity information and tribe information; The second classroom activity index includes: a comprehensive activity series index, a college series activity index and a tribal series activity index; The student network behavior information includes: the duration and number of times students visit ideological and political, entertainment video, game and educational content online; The university information includes: the name of the university, the number of students in the university, the type of university, the level of the university and the region where the university is located; The student’s personal information includes: the student’s name, the university to which the student belongs, the number of the second-classroom activity in which the student participated, the number of growth records the student obtained, the second-classroom credits the student obtained, and the role the student played in the activity course; The second classroom activity information includes: the second classroom activity number, activity name, activity time, activity type, the university hosting the activity, the tribe hosting the activity, the maximum number of registrants for the activity, the actual number of registrants for the activity, the actual number of sign-ins for the activity, the activity score and the activity duration; The tribal information includes: tribal name and tribal type; The comprehensive activity series index includes: comprehensive activity index, index of each activity type and index of each sub-region activity; The series of university activity indexes include: activity index of each university, activity index of each university type and activity index of each university level; The tribal series activity index includes: activity index of each tribal type.

3. The method for predicting the demand for second-classroom activities in colleges and universities according to claim 1 is characterized in that: Determining the weight of each evaluation index includes: Quantitative calculation of each evaluation index is performed on the extracted information related to the second classroom activities corresponding to the reference period to obtain the value of each evaluation index; The obtained values ​​of each evaluation index are used as sample data of the corresponding second classroom activity index; Using the sample data of the corresponding second classroom activity index and the sample data of the second classroom activity index similar to the corresponding second classroom activity index, a sample set is constructed; The entropy weight method is used to perform weight analysis on the sample set to obtain the weight of each evaluation index.

4. The method for predicting the demand for second-classroom activities in colleges and universities according to claim 3 is characterized in that: The method for identifying the second classroom activity index that is similar to the corresponding second classroom activity index comprises: If the second classroom activity index to be identified and the corresponding second classroom activity index are both college type activity indexes, college level activity indexes, tribe type activity indexes, activity type indexes or district activity indexes, then the second classroom activity index to be identified is similar to the corresponding second classroom activity index.

5. The method for predicting the demand for second-classroom activities in colleges and universities according to claim 1 is characterized in that: Determining the time series of scores of each evaluation index includes: Quantitative calculation of each evaluation index is performed on the extracted information related to the second classroom activities corresponding to each period of the reference period, and the values ​​of each evaluation index corresponding to each period of the reference period are obtained; Based on the values ​​of the evaluation indicators, determining the value distribution of each evaluation indicator; The values ​​of the first preset quantile, the second preset quantile and the median in the numerical distribution of each evaluation index are taken as the minimum value, the maximum value and the mean value of each evaluation index; The values ​​of each evaluation index that are outside the minimum and maximum value intervals of each evaluation index are identified as abnormal values ​​of each evaluation index; Replace the abnormal values ​​in the values ​​of each evaluation index corresponding to each period of the reference period with the corresponding mean value; Based on the maximum and minimum values ​​of each evaluation index, the frontier distance method is used to score the replaced values ​​of each evaluation index corresponding to each period of the reference period; The time series of the scores of the evaluation index values ​​corresponding to the time periods of the reference cycle are used to generate the time series of the scores of the evaluation index values.

6. The method for predicting the demand for second-classroom activities in colleges and universities according to any one of claims 1, 3 or 5, characterized in that: The evaluation indicators include: average number of activities, average activity duration, average registration rate, average sign-in rate, average activity score, average number of activity types, average number of activities participated in per capita, average activity duration per capita, average number of growth records per capita, average number of activity types registered per capita, proportion of people without growth records, ideological and moral level, and proportion of ideological guidance type activities; The demand for second-classroom activities in the college includes: the demand for various types of second-classroom activities held by various tribes in the college.

7. A device for predicting the demand for second-classroom activities in colleges and universities, characterized in that: The device comprises: The first determination module is used to analyze the second classroom activity related information of the given area in each time period of the reference period to determine the second classroom activity index time series sequence, which includes: for each second classroom activity index, extracting the second classroom activity related information associated with it from the second classroom activity related information of the given area in each time period of the reference period; based on the extracted second classroom activity related information, determining the weight of each evaluation indicator and the time series sequence of each evaluation indicator score; processing the weight of each evaluation indicator and the time series sequence of each evaluation indicator score in a weighted average manner to obtain the corresponding second classroom activity index time series sequence; A definition module is used to use the time series of the second classroom activity index and the student network behavior information of the given area in each period of the reference period as the second classroom activity features of the given area in the period to be predicted; A second determination module is used to determine the second classroom activity demand of all colleges and universities in a given area during a prediction period based on the second classroom activity characteristics and the second classroom activity demand prediction model; The reference period is a period consisting of a plurality of consecutive periods before the period to be predicted; The second classroom activity demand prediction model is obtained by training the second classroom activity characteristics of a given area in a historical period and the actual arrangement of second classroom activities of colleges and universities in a given area in a historical period.

8. 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 steps of the method for predicting the demand for second classroom activities in colleges and universities as described in any one of claims 1 to 6 are implemented.

9. 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 steps of the method for predicting the demand for second classroom activities in colleges and universities as described in any one of claims 1 to 6 are implemented.

Images