CN112085541A - User demand analysis method and device based on browsing consumption time series data - Google Patents
User demand analysis method and device based on browsing consumption time series data Download PDFInfo
- Publication number
- CN112085541A CN112085541A CN202011030181.4A CN202011030181A CN112085541A CN 112085541 A CN112085541 A CN 112085541A CN 202011030181 A CN202011030181 A CN 202011030181A CN 112085541 A CN112085541 A CN 112085541A
- Authority
- CN
- China
- Prior art keywords
- data
- user
- negative sample
- positive
- transaction
- 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.)
- Pending
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 56
- 230000006399 behavior Effects 0.000 claims abstract description 155
- 239000011159 matrix material Substances 0.000 claims abstract description 52
- 238000000034 method Methods 0.000 claims abstract description 46
- 238000012549 training Methods 0.000 claims abstract description 32
- 238000012545 processing Methods 0.000 claims abstract description 29
- 238000013528 artificial neural network Methods 0.000 claims abstract description 20
- 238000013135 deep learning Methods 0.000 claims abstract description 16
- 230000007787 long-term memory Effects 0.000 claims abstract description 14
- 230000006403 short-term memory Effects 0.000 claims abstract description 14
- 238000012544 monitoring process Methods 0.000 claims abstract description 7
- 239000013598 vector Substances 0.000 claims description 38
- 230000008859 change Effects 0.000 claims description 28
- 230000015654 memory Effects 0.000 claims description 27
- 238000004590 computer program Methods 0.000 claims description 16
- 238000007781 pre-processing Methods 0.000 claims description 15
- 238000012360 testing method Methods 0.000 claims description 15
- 238000003860 storage Methods 0.000 claims description 11
- 230000004931 aggregating effect Effects 0.000 claims description 7
- 238000013507 mapping Methods 0.000 claims description 7
- 238000010276 construction Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims 2
- 210000002569 neuron Anatomy 0.000 description 18
- 230000006870 function Effects 0.000 description 17
- 238000010586 diagram Methods 0.000 description 16
- 230000008569 process Effects 0.000 description 10
- 238000004891 communication Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 5
- 239000000872 buffer Substances 0.000 description 4
- 238000012216 screening Methods 0.000 description 4
- 230000001276 controlling effect Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000008034 disappearance Effects 0.000 description 3
- 238000004880 explosion Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 230000000717 retained effect Effects 0.000 description 2
- 230000001413 cellular effect Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000000875 corresponding effect Effects 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000009021 linear effect Effects 0.000 description 1
- 238000007477 logistic regression Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0203—Market surveys; Market polls
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/02—Banking, e.g. interest calculation or account maintenance
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Economics (AREA)
- Marketing (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Technology Law (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a user demand analysis method and a device based on browsing consumption time series data, wherein the method comprises the following steps: acquiring historical behavior data of a user in a monitoring period, and respectively intercepting the behavior data of the user in a preset period according to the occurrence condition of transaction behaviors of the user in the historical behavior data to construct positive sample data and negative sample data; performing numerical processing on the positive sample data and the negative sample data to generate a numerical sequence, and performing vectorization processing on the numerical sequence to generate a positive sample set matrix and a negative sample set matrix; training and verifying a learning model constructed based on a deep learning long-term and short-term memory neural network through the positive and negative sample set matrix to obtain a demand analysis model; and acquiring behavior data to be tested of the user, and acquiring user requirements according to the behavior data to be tested and the requirement analysis model.
Description
Technical Field
The invention relates to the field of big data, in particular to a user demand analysis method and device based on browsing consumption time series data.
Background
With the rapid development of emerging scientific technologies represented by big data, cloud computing, artificial intelligence and the like and the wide application of the emerging scientific technologies in the financial field, banks increasingly pay more attention to the fine development of product marketing. How to further enhance the effective identification of user requirements and user characteristics by utilizing accurate account historical data, user browsing records and other information related to consumption behaviors becomes an important subject of banking.
The internet is the most convenient channel for obtaining information at present, and a financial institution can effectively identify the requirements of a user by analyzing information such as transaction behaviors, browsing behaviors, clicking behaviors, searching behaviors and the like of the user, so that the degree of the requirements of the user on products is intelligently judged according to the requirements of the user; through deep mining of historical search records and click behavior data of the user, the association mode required by the user is found, and the marketing success rate of product pushing can be effectively improved. Therefore, with the rapid development of informatization, mining and analyzing mass user browsing consumption record data become more and more research hotspots, and more methods are proposed and applied to the analysis and prediction of the user demand on products, such as a Logistic regression method, a decision tree, a random forest method and the like; however, due to the technical difficulties of huge data volume, complex processing and the like, most researches still qualitatively analyze influence factors to construct theoretical models, and the performance is more or less limited. In summary, a high-precision user requirement analysis method is needed in the industry.
Disclosure of Invention
The invention aims to provide a user demand analysis method and device based on browsing consumption time series data, which are used for providing user demand analysis with high precision and remarkable effect and providing effective data support for follow-up research.
To achieve the above object, the method for analyzing user requirements based on browsing consumption time series data provided by the present invention specifically comprises: acquiring historical behavior data of a user in a monitoring period, and respectively intercepting the behavior data of the user in a preset period according to the occurrence condition of transaction behaviors of the user in the historical behavior data to construct positive sample data and negative sample data; performing numerical processing on the positive sample data and the negative sample data to generate a numerical sequence, and performing vectorization processing on the numerical sequence to generate a positive sample set matrix and a negative sample set matrix; training and verifying a learning model constructed based on a deep learning long-term and short-term memory neural network through the positive and negative sample set matrix to obtain a demand analysis model; and acquiring behavior data to be tested of the user, and acquiring user requirements according to the behavior data to be tested and the requirement analysis model.
In the above method for analyzing user demand based on browsing consumption time series data, preferably, the step of respectively intercepting behavior data of the user in a predetermined period according to occurrence conditions of transaction behaviors of the user in the historical behavior data to construct positive sample data and negative sample data includes: when the historical behavior data contains a transaction behavior, intercepting first behavior data in a preset first period before the transaction occurrence time according to the transaction occurrence time of the transaction behavior, and generating positive sample data according to the first behavior data; and when the user does not have a transaction behavior in the preset second period in the historical behavior data, intercepting second behavior data of the user in a preset third period, and generating negative sample data according to the second behavior data.
In the above user demand analysis method based on browsing consumption time series data, preferably, when the historical behavior data includes a transaction behavior, intercepting, according to the transaction occurrence time of the transaction behavior, first behavior data within a predetermined first period before the transaction occurrence time includes: when the historical behavior data comprises a plurality of transaction behaviors, taking the transaction occurrence time of the Nth transaction behavior as the starting time; and intercepting the first behavior data in a preset first period after the preset fourth period is pushed forward by taking the starting time as a starting point.
In the above method for analyzing user demand based on browsing consumption time series data, preferably, when the user does not have a transaction behavior within a predetermined second period in the historical behavior data, intercepting the second behavior data of the user within a predetermined third period includes: the third period is less than the second period, and the start time and the end time are different.
In the above method for analyzing user demand based on browsing consumption time series data, before performing a numerical processing on the positive sample data and the negative sample data to generate a numerical sequence, the method further includes: and acquiring website browsing records in the positive sample data and the negative sample data, and eliminating irrelevant websites according to the correlation between the website browsing records and the transaction behaviors to obtain associated website character strings.
In the above method for analyzing user demand based on browsing consumption time series data, preferably, the performing a numerical processing on the positive sample data and the negative sample data to generate a numerical sequence includes: numbering the associated website character strings, and constructing an associated dictionary of the website character strings and the website numbers; and generating a numerical sequence according to the associated dictionary and the associated website character string.
In the above method for analyzing user demand based on browsing consumption time series data, preferably, the vectorizing the numerical sequence to generate a positive and negative sample set matrix includes: and mapping the numerical value sequence into a multi-dimensional vector according to preset unit browsing duration to obtain a positive and negative sample set matrix.
In the above method for analyzing user demand based on browsing consumption time series data, preferably, the vectorizing the numerical sequence to generate a positive and negative sample set matrix includes: acquiring user account amount change data in the positive sample data and the negative sample data, and constructing a unit duration amount change vector by taking preset duration as a unit according to the user account amount change data; and aggregating the unit time length amount change vectors to generate a positive and negative sample set matrix.
In the above user demand analysis method based on browsing consumption time series data, preferably, the obtaining of the demand analysis model by training and verifying the learning model constructed based on the deep learning long and short term memory neural network through the positive and negative sample set matrix includes: dividing the positive and negative sample set matrix into a training set and a test set according to a preset proportion; and training the learning model through the training set, and verifying the trained learning model through the test set to obtain a demand analysis model.
The invention also provides a user demand analysis device based on browsing consumption time series data, which comprises a sample acquisition module, a preprocessing module, a model construction module and an analysis module; the sample acquisition module is used for acquiring historical behavior data of the user in a monitoring period, and respectively intercepting the behavior data of the user in a preset period according to the occurrence condition of transaction behaviors of the user in the historical behavior data to construct positive sample data and negative sample data; the preprocessing module is used for performing numerical processing on the positive sample data and the negative sample data to generate a numerical value sequence, and performing vectorization processing on the numerical value sequence to generate a positive and negative sample set matrix; the model construction module is used for training and verifying a learning model constructed based on a deep learning long-term and short-term memory neural network through the positive and negative sample set matrix to obtain a demand analysis model; the analysis module is used for acquiring behavior data to be tested of a user and obtaining user requirements according to the behavior data to be tested and the requirement analysis model.
In the above user demand analysis device based on browsing consumption time series data, preferably, the sample collection module includes an extraction unit, and the extraction unit is configured to, when the historical behavior data includes a transaction behavior, intercept first behavior data within a predetermined first period before a transaction occurrence time according to the transaction occurrence time of the transaction behavior, and generate positive sample data according to the first behavior data; and intercepting second behavior data of the user in a preset third period when the user does not have a transaction behavior in the preset second period in the historical behavior data, and generating negative sample data according to the second behavior data.
In the above user demand analysis device based on browsing consumption time series data, preferably, the preprocessing module further includes a screening unit, where the screening unit is configured to obtain website browsing records in the positive sample data and the negative sample data, and remove unrelated websites according to a correlation between the website browsing records and the transaction behavior, so as to obtain a related website character string.
In the above user demand analysis apparatus based on browsing consumption time series data, preferably, the preprocessing module includes a sequence unit and a vector unit; the sequence unit is used for numbering the associated website character strings, constructing an associated dictionary of the website character strings and website numbers, and generating a numerical sequence according to the associated dictionary and the associated website character strings; and the vector unit is used for mapping the numerical value sequence into a multi-dimensional vector according to preset unit browsing duration to obtain a positive and negative sample set matrix.
In the above user demand analysis apparatus based on browsing consumption time series data, preferably, the preprocessing module includes: acquiring user account amount change data in the positive sample data and the negative sample data, and constructing a unit duration amount change vector by taking preset duration as a unit according to the user account amount change data; and aggregating the unit time length amount change vectors to generate a positive and negative sample set matrix.
In the above user demand analysis apparatus based on browsing consumption time series data, preferably, the model building module includes: dividing the positive and negative sample set matrix into a training set and a test set according to a preset proportion; and training the learning model through the training set, and verifying the trained learning model through the test set to obtain a demand analysis model.
The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method when executing the computer program.
The present invention also provides a computer-readable storage medium storing a computer program for executing the above method.
The invention has the beneficial technical effects that: accurately analyzing the requirements of the user by combining a long-term and short-term memory neural network depth model through behavior rules contained in time sequence data such as massive user browsing records and historical consumption information; the method can process and browse consumption time sequence data, can effectively prevent the problem of convergence caused by gradient disappearance or explosion, and is very obvious in analysis effect.
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 application, illustrate embodiment(s) of the invention and together with the description serve to explain the principles of the invention. In the drawings:
fig. 1 is a schematic flowchart illustrating a method for analyzing a user requirement based on browsing consumption time series data according to an embodiment of the present invention;
fig. 2 is a schematic flow chart of positive and negative sample data acquisition according to an embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating a positive sample fetching rule according to an embodiment of the present invention;
FIG. 4 is a diagram illustrating negative sample access rules according to an embodiment of the present invention;
fig. 5 is a schematic diagram illustrating a generation process of a positive and negative sample set matrix according to an embodiment of the present invention;
fig. 6 is a schematic diagram illustrating a generation process of a positive and negative sample set matrix according to an embodiment of the present invention;
FIG. 7 is a schematic structural diagram of an LSTM model according to an embodiment of the present invention;
FIG. 8 is a schematic diagram of a neuron state provided by an embodiment of the present invention;
FIG. 9 is a schematic structural diagram of a user demand analysis apparatus based on browsing consumption time series data according to an embodiment of the present invention;
fig. 10 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the present invention is described in further detail below with reference to the embodiments and the accompanying drawings. The exemplary embodiments and descriptions of the present invention are provided to explain the present invention, but not to limit the present invention.
Additionally, the steps illustrated in the flow charts of the figures may be performed in a computer system such as a set of computer-executable instructions and, although a logical order is illustrated in the flow charts, in some cases, the steps illustrated or described may be performed in an order different than here.
Referring to fig. 1, the method for analyzing user requirements based on browsing consumption time series data provided by the present invention specifically includes:
s101, acquiring historical behavior data of a user in a monitoring period, and respectively intercepting the behavior data of the user in a preset period according to the occurrence condition of transaction behaviors of the user in the historical behavior data to construct positive sample data and negative sample data;
s102, carrying out numerical processing on the positive sample data and the negative sample data to generate a numerical sequence, and carrying out vectorization processing on the numerical sequence to generate a positive sample set matrix and a negative sample set matrix;
s103, training and verifying a learning model constructed based on a deep learning long-term and short-term memory neural network through the positive and negative sample set matrix to obtain a demand analysis model;
s104, acquiring to-be-detected behavior data of the user, and obtaining user requirements according to the to-be-detected behavior data and the requirement analysis model.
In the embodiment, the invention provides a user purchase demand analysis method based on deep learning by combining a long-term and short-term memory neural network depth model based on behavior rules contained in time sequence data such as massive user browsing records, historical consumption information and the like; in recent years, deep learning is one of the technologies with the highest attention in the field of machine learning and artificial intelligence, and deep learning models make great breakthrough in many fields by virtue of strong expression capacity and network structures. According to the method, a deep learning technology is applied to a user demand analysis service scene, a long-term and short-term memory neural network deep model is integrated and constructed based on multi-platform user browsing records and account transaction information time sequence data, the demand opportunity of a user on a product is predicted and analyzed, and the product is accurately pushed to the user, so that the cost of manpower and material resources is reduced, the satisfaction degree of the user can be improved, and the retention rate of the user is increased.
Referring to fig. 2, in an embodiment of the present invention, intercepting behavior data of a user in a predetermined period according to occurrence conditions of transaction behaviors of the user in the historical behavior data to construct positive sample data and negative sample data respectively includes:
s201, when the historical behavior data contains a transaction behavior, intercepting first behavior data in a preset first period before the transaction occurrence time according to the transaction occurrence time of the transaction behavior, and generating positive sample data according to the first behavior data;
s202, when the user does not have a transaction behavior in the preset second period in the historical behavior data, intercepting second behavior data of the user in a preset third period, and generating negative sample data according to the second behavior data.
In the above embodiment, when the historical behavior data includes a transaction behavior, intercepting, according to the transaction occurrence time of the transaction behavior, first behavior data within a predetermined first period before the transaction occurrence time includes: when the historical behavior data comprises a plurality of transaction behaviors, taking the transaction occurrence time of the Nth transaction behavior as the starting time; and intercepting the first behavior data in a preset first period after the preset fourth period is pushed forward by taking the starting time as a starting point. In another embodiment, when the user does not have a transaction behavior within a predetermined second period in the historical behavior data, intercepting the second behavior data of the user within a predetermined third period comprises: the third period is less than the second period, and the start time and the end time are different.
In actual work, the user browsing consumption time series data is data which changes continuously along with the time, and the data can be mainly divided into two types: 1. a user accesses browsing records of a PC (personal computer) end and a mobile phone APP (application); 2. real-time data of the amount of the bank account of the user; therefore, a positive and negative sample set of the user browsing consumption time series data can be constructed by confirming the positive and negative labels and the input data range. Thus, in this embodiment, the positive sample label setting rule may be set to: setting a positive sample label for a user with purchasing behavior in an observation period; if multiple purchases occur, the first purchase is the date the product was purchased. The positive sample data input is: the date of purchasing the product by the user is pushed forward by 15 days, and the behavior data of the previous 30 days is taken as the input of the positive sample of the model, which can be specifically shown in fig. 3. The 15-day interval is set because if the data of 30 days before the date of purchasing the product by the user is directly taken, the last day is taken, and only 1 day of purchasing behavior is taken, the model actually changes to 'whether the user purchases the product on the next day is predicted by the data of the last 30 days', and the difference exists between the data and the fact; the goal here is to predict the probability that a user will demand a product within 30 days in the future, while 30 days include day 1 and day 30 in the future, with an average of 15 days, so the time interval is chosen to be 15 days. On the other side, the negative example label setting rule is: during the observation period, the user without any purchase record sets the positive sample tab. The label of the negative sample is better determined, and the negative sample data input provided by the embodiment adopts a scheme of taking numbers from the middle: the data of the middle 30 days in the observation period are selected as the input of the model negative sample, as shown in fig. 4. The reason why the user takes the number from the middle, not from the end, is that the result that the user has not purchased the product after the number taking period is definitely obtained, which can confirm that the user is a negative sample; if only the number is taken from the end, it is not excluded that the user suddenly purchases the product after the observation period, and thus the user should be classified as a positive sample; taking the number from the middle is chosen as the input for the negative examples of the dynamic model based on a robust principle.
In an embodiment of the present invention, before performing a numerical processing on the positive sample data and the negative sample data to generate a numerical sequence, the method further includes: and acquiring website browsing records in the positive sample data and the negative sample data, and eliminating irrelevant websites according to the correlation between the website browsing records and the transaction behaviors to obtain associated website character strings. As such, subsequent digitizing the positive sample data and the negative sample data to generate a sequence of numerical values may comprise: numbering the associated website character strings, and constructing an associated dictionary of the website character strings and the website numbers; and generating a numerical sequence according to the associated dictionary and the associated website character string. Based on the foregoing embodiment, in another embodiment of the present invention, the vectorizing the numerical sequence to generate a positive and negative sample set matrix includes: mapping the numerical value sequence into a multi-dimensional vector according to preset unit browsing duration to obtain a positive and negative sample set matrix; reference is made in particular to fig. 5.
In actual work, browsing records of a user accessing a PC (personal computer) end and a mobile phone APP (application) are composed of a plurality of websites, each website is of a character string type, and the matrixing process of positive and negative sample sets comprises the following steps:
1. and (4) data cleaning, namely removing irrelevant websites from the browsing records of the sample set user, and reserving website character strings relevant to product purchase of the user.
2. Digitizing the user browsing records, numbering all website character strings in the user browsing records, and constructing a website character string-website number dictionary. In this way, the browsing history of each user can be represented as a numerical sequence consisting of a string of numbers of the purchase visit web sites.
3. And vectorizing the numerical sequence of the browsing record numbers of the user, and mapping the browsing record number sequence into a multi-dimensional vector. The dimension of the multidimensional vector is recorded as L, and the L value can be set by integrating the maximum website sequence length of actual user browsing records in the observation period and the calculation cost condition of the LSTM model. If the length of the sequence of the website browsed and recorded by the user exceeds L, only the first L website numbers are reserved, and if the length of the sequence of the website browsed and recorded by the user is less than L, 0 is used for filling. Therefore, all the user browsing records are vectors with L website numbers, each website number is converted into an N-dimensional vector, L multiplied by N-dimensional matrixes representing the user visiting records can be obtained, and each matrix is provided with a positive sample label and a negative sample label.
Referring to fig. 6, in another embodiment of the present invention, the vectorizing the numerical sequence to generate a positive and negative sample set matrix includes: acquiring user account amount change data in the positive sample data and the negative sample data, and constructing a unit duration amount change vector by taking preset duration as a unit according to the user account amount change data; and aggregating the unit time length amount change vectors to generate a positive and negative sample set matrix. In actual work, real-time data of the amount of a bank account of a user is numerical data, positive and negative samples can be directly matrixed, and the specific flow is as follows:
1. and performing data slicing on the real-time change data of the user bank account amount by taking the day as a unit, and constructing a change vector of the user single-day account amount.
2. Aggregating the change vectors of the account amount of the user on a single day, and constructing a time sequence matrix of the account amount change of the user on 30 days, wherein each matrix is provided with a positive sample label and a negative sample label.
It should be noted that fig. 5 and fig. 6 and the corresponding embodiments thereof can be combined in practical work, and those skilled in the art can select the combination according to practical needs, and the invention is not further limited herein.
In an embodiment of the present invention, the obtaining of the demand analysis model by training and verifying the learning model constructed based on the deep learning long-term and short-term memory neural network through the positive and negative sample set matrix includes: dividing the positive and negative sample set matrix into a training set and a test set according to a preset proportion; and training the learning model through the training set, and verifying the trained learning model through the test set to obtain a demand analysis model. In actual work, a positive and negative sample set of user browsing consumption time series data can be divided into a training set and a testing set, wherein the training set is used for training and constructing a user purchase demand prediction model, and the testing set is used for testing and checking the prediction effect of the model; this flow may be implemented using existing techniques and will not be illustrated in detail herein.
In an embodiment of the present invention, when the learning model is actually applied, a long-short term memory neural network LSTM model is mainly used, although those skilled in the art may select other learning models according to actual needs, and the present invention is not limited herein. The LSTM model provided by the invention can add or delete information to or from the state of the neurons in the neural network mainly through a unique gate structure, thereby interactively controlling the state of the neurons, changing the information carried by the state of the neurons, and protecting and controlling the state of the neurons; the structure is shown in fig. 7.
Xt-1、Xt、Xt+1The user representing the input model consumes the browsing timing data. h ist-1、ht、ht+1The hidden state of the recurrent neural network RNN is represented, and may be regarded as an intermediate vector output by each recurrent node a and stored as state information. The hidden state is used as the output of the previous node and is transmitted to the next node as the input, and the output value h of each cycletAre all in contact with XtAnd ht-1Is correlated. Each loop node a of the RNN will have an output variable Y for the previous time sequence variable X, Y at each instant of time being a function (possibly Sigmoid or softmax function) representation of the current output state h. The LSTM adds a core vector C on the basis of RNNtOne horizontal line through the entire structure is called neuron state information. As shown in fig. 8:
the neuron state runs through the whole chain structure, has small linear action and mainly plays a role of transmitting information which comprises information forgotten when passing through each node A and newly input information. LSTM proposes the concept of multi-layer gate control, namely a forgetting gate, an input gate and an output gate; the core process of the LSTM model is as follows:
1. forget the door: the information that decides should be forgotten by the neuron, consists of a Sigmoid layer. Forget gate to read input h of last hidden state layert-1And input X of the current time pointtIn neuronal state Ct-1Output value ft，ftThe value range is [0, 1 ]]"1" represents "completely retained information" and "0" represents "completely forgotten information", and the specific calculation formula is as follows:
ft＝σ(Wf·[ht-1，Xt]+bf)；
wherein, WfIs the weight matrix of the forgetting gate, [ h ]t-1，Xt]Representing the concatenation of two vectors into a longer vector, bfIs the bias term for the forgetting gate, σ is the Sigmoid function.
2. An input gate: the information that determines that should be retained by the neuron consists of two parts. The first part determines the value to be updated through a Sigmoid layer, and the second part creates an intermediate value through a tanh layer
Add to Current neuron State CtPerforming the following steps; the specific calculation formula is as follows:
it＝σ(Wi·[ht-1，Xt]+bi)；
wherein, Wi、WCIs the weight matrix of the input gate, [ h ]t-1，Xt]Representing the concatenation of two vectors into a longer vector, bi、bCIs the bias term of the input gate, σ is the Sigmoid function, and tanh is the tanh function.
3. Neuron state updating: combining forget gate and input gate can update neuron state and update old neuron state Ct-1To new neuron state Ct. Multiplying old neuron state by forgetting gate output value ftBefore forgetting, the information to be forgotten is determined, and then the product of the two parts of the input gate is added, so that the state of the neuron is updated to the latest CtA state value; the specific calculation formula is as follows:
4. an input gate: the output information of the decision neural network consists of two parts. The first part determines which part of neuron states are output by using a Sigmoid layer, and the second part processes the current states by using a tanh layer and then multiplies the current states by the outputs of the first part, so that the part which should be output is ensured to be output; the specific calculation formula is as follows:
ot＝σ(Wo·[ht-1，Xt]+bo)；
ht＝ot·tanh(Ct)；
wherein, WoIs a weight matrix of output gates, [ h ]t-1，Xt]Representing the concatenation of two vectors into a longer vector, boIs the offset term of the output gate, σ is the Sigmoid function, and tanh is the tanh function. Final output htComputing the result o from the first parttAnd neuron state value CtAnd (4) jointly determining.
Through the four steps, the positive and negative sample sets of the user browsing consumption time series data can be effectively trained and predicted, the user purchasing demand prediction LSTM model obtained through continuous iterative optimization in the training process can provide a prediction score for the demand of the user for purchasing products in 30 days in the future, and the higher the prediction score is, the higher the demand intensity of the user for the products in 30 days in the future is, and the higher the purchasing probability is.
According to the method, the purchasing demand of the user on the product in a certain period in the future is recognized by constructing three links of a positive sample set and a negative sample set of a user browsing consumption time sequence, matrixing the positive sample set and the negative sample set and constructing a model based on a deep learning long-term and short-term memory neural network, so that accurate marketing of the user is realized; the LSTM model adopted by the invention not only can process and browse consumption time series data, but also can effectively prevent the convergence problem caused by gradient disappearance or explosion, and has very obvious prediction effect; meanwhile, the deep learning long-short term memory neural network technology is applied to analysis and prediction of the purchase demand of the user, time sequence data can be effectively processed, data characteristics can be automatically obtained, the quantized time sequence positive and negative sample sets are input into the model, the analysis and prediction result of the user demand can be directly output, and the deep learning long-short term memory neural network model has the advantages of high accuracy, strong generalization capability, capability of processing massive data and the like.
Referring to fig. 9, the present invention further provides a user demand analysis device based on browsing consumption time series data, where the device includes a sample collection module, a preprocessing module, a model construction module, and an analysis module; the sample acquisition module is used for acquiring historical behavior data of the user in a monitoring period, and respectively intercepting the behavior data of the user in a preset period according to the occurrence condition of transaction behaviors of the user in the historical behavior data to construct positive sample data and negative sample data; the preprocessing module is used for performing numerical processing on the positive sample data and the negative sample data to generate a numerical value sequence, and performing vectorization processing on the numerical value sequence to generate a positive and negative sample set matrix; the model construction module is used for training and verifying a learning model constructed based on a deep learning long-term and short-term memory neural network through the positive and negative sample set matrix to obtain a demand analysis model; the analysis module is used for acquiring behavior data to be tested of a user and obtaining user requirements according to the behavior data to be tested and the requirement analysis model.
In the above embodiment, the sample collection module includes an extraction unit, and the extraction unit is configured to, when the historical behavior data includes a transaction behavior, intercept, according to a transaction occurrence time of the transaction behavior, first behavior data within a predetermined first period before the transaction occurrence time, and generate positive sample data according to the first behavior data; and intercepting second behavior data of the user in a preset third period when the user does not have a transaction behavior in the preset second period in the historical behavior data, and generating negative sample data according to the second behavior data.
In an embodiment of the present invention, the preprocessing module further includes a screening unit, where the screening unit is configured to obtain a website browsing record in the positive sample data and the negative sample data, and remove unrelated websites according to a correlation between the website browsing record and the transaction behavior to obtain an associated website character string. Further, in another embodiment of the present invention, the preprocessing module comprises a sequence unit and a vector unit; the sequence unit is used for numbering the associated website character strings, constructing an associated dictionary of the website character strings and website numbers, and generating a numerical sequence according to the associated dictionary and the associated website character strings; and the vector unit is used for mapping the numerical value sequence into a multi-dimensional vector according to preset unit browsing duration to obtain a positive and negative sample set matrix.
In an embodiment of the present invention, the preprocessing module includes: acquiring user account amount change data in the positive sample data and the negative sample data, and constructing a unit duration amount change vector by taking preset duration as a unit according to the user account amount change data; and aggregating the unit time length amount change vectors to generate a positive and negative sample set matrix.
In an embodiment of the present invention, the model building module includes: dividing the positive and negative sample set matrix into a training set and a test set according to a preset proportion; and training the learning model through the training set, and verifying the trained learning model through the test set to obtain a demand analysis model.
The specific implementation of each module in the user requirement analysis apparatus based on browsing consumption time series data provided by the present invention has been described in detail in the foregoing embodiments, and thus, detailed description is omitted here.
The invention has the beneficial technical effects that: accurately analyzing the requirements of the user by combining a long-term and short-term memory neural network depth model through behavior rules contained in time sequence data such as massive user browsing records and historical consumption information; the method can process and browse consumption time sequence data, can effectively prevent the problem of convergence caused by gradient disappearance or explosion, and is very obvious in analysis effect.
The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method when executing the computer program.
The present invention also provides a computer-readable storage medium storing a computer program for executing the above method.
As shown in fig. 10, the electronic device 600 may further include: communication module 110, input unit 120, audio processing unit 130, display 160, power supply 170. It is noted that the electronic device 600 does not necessarily include all of the components shown in FIG. 10; furthermore, the electronic device 600 may also comprise components not shown in fig. 10, which may be referred to in the prior art.
As shown in fig. 10, the central processor 100, sometimes referred to as a controller or operational control, may include a microprocessor or other processor device and/or logic device, the central processor 100 receiving input and controlling the operation of the various components of the electronic device 600.
The memory 140 may be, for example, one or more of a buffer, a flash memory, a hard drive, a removable media, a volatile memory, a non-volatile memory, or other suitable device. The information relating to the failure may be stored, and a program for executing the information may be stored. And the central processing unit 100 may execute the program stored in the memory 140 to realize information storage or processing, etc.
The input unit 120 provides input to the cpu 100. The input unit 120 is, for example, a key or a touch input device. The power supply 170 is used to provide power to the electronic device 600. The display 160 is used to display an object to be displayed, such as an image or a character. The display may be, for example, an LCD display, but is not limited thereto.
The memory 140 may be a solid state memory such as Read Only Memory (ROM), Random Access Memory (RAM), a SIM card, or the like. There may also be a memory that holds information even when power is off, can be selectively erased, and is provided with more data, an example of which is sometimes called an EPROM or the like. The memory 140 may also be some other type of device. Memory 140 includes buffer memory 141 (sometimes referred to as a buffer). The memory 140 may include an application/function storage section 142, and the application/function storage section 142 is used to store application programs and function programs or a flow for executing the operation of the electronic device 600 by the central processing unit 100.
The memory 140 may also include a data store 143, the data store 143 for storing data, such as contacts, digital data, pictures, sounds, and/or any other data used by the electronic device. The driver storage portion 144 of the memory 140 may include various drivers of the electronic device for communication functions and/or for performing other functions of the electronic device (e.g., messaging application, address book application, etc.).
The communication module 110 is a transmitter/receiver 110 that transmits and receives signals via an antenna 111. The communication module (transmitter/receiver) 110 is coupled to the central processor 100 to provide an input signal and receive an output signal, which may be the same as in the case of a conventional mobile communication terminal.
Based on different communication technologies, a plurality of communication modules 110, such as a cellular network module, a bluetooth module, and/or a wireless local area network module, may be provided in the same electronic device. The communication module (transmitter/receiver) 110 is also coupled to a speaker 131 and a microphone 132 via an audio processor 130 to provide audio output via the speaker 131 and receive audio input from the microphone 132 to implement general telecommunications functions. Audio processor 130 may include any suitable buffers, decoders, amplifiers and so forth. In addition, an audio processor 130 is also coupled to the central processor 100, so that recording on the local can be enabled through a microphone 132, and so that sound stored on the local can be played through a speaker 131.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (17)
1. A user demand analysis method based on browsing consumption time series data is characterized by comprising the following steps:
acquiring historical behavior data of a user in a monitoring period, and respectively intercepting the behavior data of the user in a preset period according to the occurrence condition of transaction behaviors of the user in the historical behavior data to construct positive sample data and negative sample data;
performing numerical processing on the positive sample data and the negative sample data to generate a numerical sequence, and performing vectorization processing on the numerical sequence to generate a positive sample set matrix and a negative sample set matrix;
training and verifying a learning model constructed based on a deep learning long-term and short-term memory neural network through the positive and negative sample set matrix to obtain a demand analysis model;
and acquiring behavior data to be tested of the user, and acquiring user requirements according to the behavior data to be tested and the requirement analysis model.
2. The method of claim 1, wherein the step of intercepting behavior data of the user in a predetermined period according to occurrence of a transaction behavior of the user in the historical behavior data to construct positive sample data and negative sample data comprises:
when the historical behavior data contains a transaction behavior, intercepting first behavior data in a preset first period before the transaction occurrence time according to the transaction occurrence time of the transaction behavior, and generating positive sample data according to the first behavior data;
and when the user does not have a transaction behavior in the preset second period in the historical behavior data, intercepting second behavior data of the user in a preset third period, and generating negative sample data according to the second behavior data.
3. The method for analyzing user demand based on browsing consumption time series data according to claim 2, wherein when the historical behavior data includes a transaction behavior, intercepting the first behavior data within a predetermined first period before the transaction occurrence time according to the transaction occurrence time of the transaction behavior includes: when the historical behavior data comprises a plurality of transaction behaviors, taking the transaction occurrence time of the Nth transaction behavior as the starting time; and intercepting the first behavior data in a preset first period after the preset fourth period is pushed forward by taking the starting time as a starting point.
4. The method of claim 2, wherein intercepting the second behavior data of the user during a predetermined third period when the user does not perform a transaction during the predetermined second period in the historical behavior data comprises: the third period is less than the second period, and the start time and the end time are different.
5. The method of claim 1, wherein before performing the numerical processing on the positive sample data and the negative sample data to generate the numerical sequence, further comprising:
and acquiring website browsing records in the positive sample data and the negative sample data, and eliminating irrelevant websites according to the correlation between the website browsing records and the transaction behaviors to obtain associated website character strings.
6. The method of claim 5, wherein numerically processing the positive sample data and the negative sample data to generate a numerical sequence comprises:
numbering the associated website character strings, and constructing an associated dictionary of the website character strings and the website numbers;
and generating a numerical sequence according to the associated dictionary and the associated website character string.
7. The method of claim 6, wherein vectorizing the sequence of values to generate a positive and negative sample set matrix comprises: and mapping the numerical value sequence into a multi-dimensional vector according to preset unit browsing duration to obtain a positive and negative sample set matrix.
8. The method of claim 1, wherein vectorizing the sequence of values to generate a positive and negative sample set matrix comprises:
acquiring user account amount change data in the positive sample data and the negative sample data, and constructing a unit duration amount change vector by taking preset duration as a unit according to the user account amount change data;
and aggregating the unit time length amount change vectors to generate a positive and negative sample set matrix.
9. The method of claim 1, wherein training and verifying a learning model based on deep learning long-short term memory neural network to obtain a demand analysis model comprises:
dividing the positive and negative sample set matrix into a training set and a test set according to a preset proportion;
and training the learning model through the training set, and verifying the trained learning model through the test set to obtain a demand analysis model.
10. A user demand analysis device based on browsing consumption time series data is characterized by comprising a sample acquisition module, a preprocessing module, a model construction module and an analysis module;
the sample acquisition module is used for acquiring historical behavior data of the user in a monitoring period, and respectively intercepting the behavior data of the user in a preset period according to the occurrence condition of transaction behaviors of the user in the historical behavior data to construct positive sample data and negative sample data;
the preprocessing module is used for performing numerical processing on the positive sample data and the negative sample data to generate a numerical value sequence, and performing vectorization processing on the numerical value sequence to generate a positive and negative sample set matrix;
the model construction module is used for training and verifying a learning model constructed based on a deep learning long-term and short-term memory neural network through the positive and negative sample set matrix to obtain a demand analysis model;
the analysis module is used for acquiring behavior data to be tested of a user and obtaining user requirements according to the behavior data to be tested and the requirement analysis model.
11. The apparatus according to claim 10, wherein the sample collection module includes an extraction unit, and the extraction unit is configured to, when the historical behavior data includes a transaction behavior, intercept first behavior data within a predetermined first period before a transaction occurrence time of the transaction behavior according to the transaction occurrence time, and generate positive sample data according to the first behavior data; and intercepting second behavior data of the user in a preset third period when the user does not have a transaction behavior in the preset second period in the historical behavior data, and generating negative sample data according to the second behavior data.
12. The apparatus according to claim 10, wherein the preprocessing module further comprises a filtering unit, the filtering unit is configured to obtain website browsing records in the positive sample data and the negative sample data, and remove unrelated websites according to the correlation between the website browsing records and the transaction behavior to obtain associated website character strings.
13. The apparatus for analyzing user's demand based on browsing consumption time-series data of claim 12, wherein the preprocessing module comprises a sequence unit and a vector unit;
the sequence unit is used for numbering the associated website character strings, constructing an associated dictionary of the website character strings and website numbers, and generating a numerical sequence according to the associated dictionary and the associated website character strings;
and the vector unit is used for mapping the numerical value sequence into a multi-dimensional vector according to preset unit browsing duration to obtain a positive and negative sample set matrix.
14. The apparatus for analyzing user's demand based on browsing consumption time series data of claim 10, wherein the preprocessing module comprises: acquiring user account amount change data in the positive sample data and the negative sample data, and constructing a unit duration amount change vector by taking preset duration as a unit according to the user account amount change data; and aggregating the unit time length amount change vectors to generate a positive and negative sample set matrix.
15. The apparatus for analyzing user's demand based on browsing consumption time-series data of claim 10, wherein the model building module comprises: dividing the positive and negative sample set matrix into a training set and a test set according to a preset proportion; and training the learning model through the training set, and verifying the trained learning model through the test set to obtain a demand analysis model.
16. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of any of claims 1 to 9 when executing the computer program.
17. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program for executing the method of any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011030181.4A CN112085541A (en) | 2020-09-27 | 2020-09-27 | User demand analysis method and device based on browsing consumption time series data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011030181.4A CN112085541A (en) | 2020-09-27 | 2020-09-27 | User demand analysis method and device based on browsing consumption time series data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112085541A true CN112085541A (en) | 2020-12-15 |
Family
ID=73738636
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011030181.4A Pending CN112085541A (en) | 2020-09-27 | 2020-09-27 | User demand analysis method and device based on browsing consumption time series data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112085541A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112381595A (en) * | 2021-01-11 | 2021-02-19 | 平安科技（深圳）有限公司 | User value prediction method based on communication behavior and related equipment |
| CN112541745A (en) * | 2020-12-22 | 2021-03-23 | 平安银行股份有限公司 | User behavior data analysis method and device, electronic equipment and readable storage medium |
| CN112882934A (en) * | 2021-02-24 | 2021-06-01 | 中国工商银行股份有限公司 | Test analysis method and system based on defect growth |
| CN113138977A (en) * | 2021-04-22 | 2021-07-20 | 康键信息技术（深圳）有限公司 | Transaction conversion analysis method, device, equipment and storage medium |
| CN113393299A (en) * | 2021-06-17 | 2021-09-14 | 中国工商银行股份有限公司 | Recommendation model training method and device, electronic equipment and storage medium |
| CN113988152A (en) * | 2021-09-23 | 2022-01-28 | 北京达佳互联信息技术有限公司 | User type prediction model training method, resource allocation method, medium, and apparatus |
| CN115511546A (en) * | 2022-11-23 | 2022-12-23 | 深圳市云积分科技有限公司 | Behavior analysis method, system, equipment and readable medium for E-commerce users |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107704918A (en) * | 2017-09-19 | 2018-02-16 | 平安科技（深圳）有限公司 | Driving model training method, driver's recognition methods, device, equipment and medium |
| CN107944913A (en) * | 2017-11-21 | 2018-04-20 | 重庆邮电大学 | High potential user's purchase intention Forecasting Methodology based on big data user behavior analysis |
| CN110555717A (en) * | 2019-07-29 | 2019-12-10 | 华南理工大学 | method for mining potential purchased goods and categories of users based on user behavior characteristics |
| CN110689164A (en) * | 2019-08-26 | 2020-01-14 | 阿里巴巴集团控股有限公司 | Prediction method and system for user reduction behavior |
| CN110837683A (en) * | 2019-05-20 | 2020-02-25 | 全球能源互联网研究院有限公司 | Training and predicting method and device for prediction model of transient stability of power system |
| CN111222677A (en) * | 2019-10-22 | 2020-06-02 | 浙江运达风电股份有限公司 | Wind speed prediction method and system based on long-short term memory time neural network |
-
2020
- 2020-09-27 CN CN202011030181.4A patent/CN112085541A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107704918A (en) * | 2017-09-19 | 2018-02-16 | 平安科技（深圳）有限公司 | Driving model training method, driver's recognition methods, device, equipment and medium |
| CN107944913A (en) * | 2017-11-21 | 2018-04-20 | 重庆邮电大学 | High potential user's purchase intention Forecasting Methodology based on big data user behavior analysis |
| CN110837683A (en) * | 2019-05-20 | 2020-02-25 | 全球能源互联网研究院有限公司 | Training and predicting method and device for prediction model of transient stability of power system |
| CN110555717A (en) * | 2019-07-29 | 2019-12-10 | 华南理工大学 | method for mining potential purchased goods and categories of users based on user behavior characteristics |
| CN110689164A (en) * | 2019-08-26 | 2020-01-14 | 阿里巴巴集团控股有限公司 | Prediction method and system for user reduction behavior |
| CN111222677A (en) * | 2019-10-22 | 2020-06-02 | 浙江运达风电股份有限公司 | Wind speed prediction method and system based on long-short term memory time neural network |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112541745A (en) * | 2020-12-22 | 2021-03-23 | 平安银行股份有限公司 | User behavior data analysis method and device, electronic equipment and readable storage medium |
| CN112541745B (en) * | 2020-12-22 | 2024-04-09 | 平安银行股份有限公司 | User behavior data analysis method and device, electronic equipment and readable storage medium |
| CN112381595A (en) * | 2021-01-11 | 2021-02-19 | 平安科技（深圳）有限公司 | User value prediction method based on communication behavior and related equipment |
| CN112381595B (en) * | 2021-01-11 | 2021-06-04 | 平安科技（深圳）有限公司 | User value prediction method based on communication behavior and related equipment |
| CN112882934A (en) * | 2021-02-24 | 2021-06-01 | 中国工商银行股份有限公司 | Test analysis method and system based on defect growth |
| CN112882934B (en) * | 2021-02-24 | 2024-02-13 | 中国工商银行股份有限公司 | Test analysis method and system based on defect growth |
| CN113138977A (en) * | 2021-04-22 | 2021-07-20 | 康键信息技术（深圳）有限公司 | Transaction conversion analysis method, device, equipment and storage medium |
| CN113393299A (en) * | 2021-06-17 | 2021-09-14 | 中国工商银行股份有限公司 | Recommendation model training method and device, electronic equipment and storage medium |
| CN113988152A (en) * | 2021-09-23 | 2022-01-28 | 北京达佳互联信息技术有限公司 | User type prediction model training method, resource allocation method, medium, and apparatus |
| CN115511546A (en) * | 2022-11-23 | 2022-12-23 | 深圳市云积分科技有限公司 | Behavior analysis method, system, equipment and readable medium for E-commerce users |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112085541A (en) | User demand analysis method and device based on browsing consumption time series data | |
| US11829874B2 (en) | Neural architecture search | |
| CN111966914B (en) | Content recommendation method and device based on artificial intelligence and computer equipment | |
| CN111090756B (en) | Artificial intelligence-based multi-target recommendation model training method and device | |
| CN111667308A (en) | Advertisement recommendation prediction system and method | |
| CN111126495B (en) | Model training method, information prediction device, storage medium and equipment | |
| CN111258593B (en) | Application program prediction model building method and device, storage medium and terminal | |
| CN112905876B (en) | Information pushing method and device based on deep learning and computer equipment | |
| CN112380449B (en) | Information recommendation method, model training method and related device | |
| CN112528010B (en) | Knowledge recommendation method and device, computer equipment and readable storage medium | |
| CN116089708A (en) | Agricultural knowledge recommendation method and device | |
| CN114996486A (en) | Data recommendation method and device, server and storage medium | |
| CN117036843A (en) | Target detection model training method, target detection method and device | |
| CN110807693A (en) | Album recommendation method, device, equipment and storage medium | |
| CN117972206A (en) | Content recommendation system, electronic equipment and storage medium based on artificial intelligence | |
| CN112749238A (en) | Search ranking method and device, electronic equipment and computer-readable storage medium | |
| CN117312657A (en) | Abnormal function positioning method and device for financial application, computer equipment and medium | |
| CN116306672A (en) | Data processing method and device | |
| CN116167371A (en) | Product recommendation method and device, processor and electronic equipment | |
| CN114842382A (en) | Method, device, equipment and medium for generating semantic vector of video | |
| CN115700550A (en) | Label classification model training and object screening method, device and storage medium | |
| CN115482019A (en) | Activity attention prediction method and device, electronic equipment and storage medium | |
| CN117828193B (en) | Multi-interest semi-joint learning interest recommendation method, system, equipment and medium | |
| CN111506510B (en) | Software quality determining method and related device | |
| CN110503482B (en) | Article processing method, device, terminal 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 |