CN113626606B - Information classification method, device, electronic equipment and readable storage medium - Google Patents
Information classification method, device, electronic equipment and readable storage medium Download PDFInfo
- Publication number
- CN113626606B CN113626606B CN202111011806.7A CN202111011806A CN113626606B CN 113626606 B CN113626606 B CN 113626606B CN 202111011806 A CN202111011806 A CN 202111011806A CN 113626606 B CN113626606 B CN 113626606B
- Authority
- CN
- China
- Prior art keywords
- classification
- history information
- information set
- model
- user
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000003860 storage Methods 0.000 title claims description 18
- 238000013145 classification model Methods 0.000 claims abstract description 59
- 238000013136 deep learning model Methods 0.000 claims abstract description 58
- 238000012549 training Methods 0.000 claims abstract description 45
- 238000012216 screening Methods 0.000 claims abstract description 11
- 230000006870 function Effects 0.000 claims description 21
- 238000004590 computer program Methods 0.000 claims description 12
- 238000013528 artificial neural network Methods 0.000 claims description 8
- 238000013473 artificial intelligence Methods 0.000 abstract description 5
- 238000005516 engineering process Methods 0.000 abstract description 4
- 238000004891 communication Methods 0.000 description 9
- 238000005457 optimization Methods 0.000 description 6
- 238000013135 deep learning Methods 0.000 description 5
- 238000007726 management method Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 230000004913 activation Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 210000002569 neuron Anatomy 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/08—Insurance
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Business, Economics & Management (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Probability & Statistics with Applications (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to artificial intelligence technology, and discloses an information classification method, which comprises the following steps: dividing a user history information set into a first history information set and a second history information set by using a preset field; training the deep learning model by using the first historical information set to obtain a first classification model; carrying out category label marking on each user history information in the second history information set by using the first classification model; training the deep learning model by using the marked second historical information set and the marked first historical information set to obtain a second classification model; classifying the user information to be evaluated by using a second classification model to obtain classification probability values of a plurality of classification categories; and screening all classification categories according to the classification probability value to obtain classification results. The present invention also relates to a blockchain technique, wherein the user history information can be stored in a blockchain node. The invention also provides an information classification device, equipment and medium. The invention can improve the accuracy of information classification.
Description
Technical Field
The present invention relates to artificial intelligence technology, and in particular, to an information classification method, apparatus, electronic device, and readable storage medium.
Background
Along with the development of economy, the requirements of users on various services are higher and higher, so that in order to better serve the users, the quality and efficiency of the services are improved, the information of the users needs to be classified, and personalized services are performed on the users according to the classification result, such as: insurance appropriate for the user is recommended.
However, in the current information classification method, model training is directly performed by using data marked by a label, information classification is performed on information to be classified by using the trained model, and the marked data is difficult to obtain, so that the performance of the model is poor, and the accuracy of information classification is low.
Disclosure of Invention
The invention provides an information classification method, an information classification device, electronic equipment and a computer readable storage medium, and mainly aims to improve the accuracy of information classification.
In order to achieve the above object, the present invention provides an information classification method, including:
acquiring a user history information set, and dividing the user history information set into a first history information set and a second history information set by utilizing a preset field;
training a pre-constructed deep learning model by using the first historical information set to obtain a first classification model;
performing category label marking on each user history information in the second history information set by using the first classification model;
training the deep learning model by using the second historical information set and the first historical information set marked by the labels to obtain the second classification model;
when receiving user information to be classified, classifying the user information to be evaluated by using the second classification model to obtain classification probability values of a plurality of classification categories;
and screening all the classification categories according to the classification probability value to obtain classification results.
Optionally, the dividing the user history information set into the first history information set and the second history information set by using a preset field includes:
selecting all user history information with values of preset fields not being null values in the user history information set, and taking the values of the preset fields as class labels of corresponding user history information to obtain the first history information set;
and selecting all user history information with the value of a preset field in the user history information set as a null value to obtain the second history information set.
Optionally, the performing category label marking on each user history information in the second history information set by using the first classification model includes:
inputting each history information in the second history information set into the first classification model to obtain class probability values of a plurality of classes;
selecting a class probability value which is larger than a preset threshold value from all class probability values of each user history information to obtain a class probability value set;
judging the number of maximum values in the class probability value set;
if the number of the maximum values is not equal to 1, deleting the user history information corresponding to the category probability value set in the second history information set;
if the number of the maximum values is not equal to 1, determining the category corresponding to the maximum value in the category probability value set as the category label of the corresponding user history information.
Optionally, training the deep learning model by using the second historical information set and the first historical information set of the label mark to obtain the second classification model, including:
combining the second historical information set marked by the label with the first historical information set to obtain an updated first historical information set;
and training the deep learning model by using the updated first historical information set to obtain the second classification model.
Optionally, before the training of the pre-constructed deep learning model using the first historical information set, the method further includes:
acquiring a deep neural network frame;
constructing a feature input layer in the deep neural network framework;
constructing a weight initialization layer after the feature input layer;
establishing a plurality of full connection layers after the characteristic input layer;
constructing batch standardization layers and discarding layers among the plurality of full connection layers;
and constructing a decision output layer after the plurality of full-connection layers to obtain the deep learning model.
Optionally, before the training of the pre-constructed deep learning model using the first historical information set, the method further includes:
selecting user history information in the first history information set, inputting the user history information into the deep learning model, and extracting an output value of a decision output layer in the deep learning model to obtain a classification predicted value;
determining a classification authenticity value according to each user history information category label in the first history information set;
calculating a preset loss function according to the classification reality value and the classification predicted value to obtain a loss value;
when the loss value is greater than or equal to a preset loss threshold value, adjusting model parameters of the deep learning model according to the loss value, and returning the user history information in the selected first history information set to input the deep learning model;
and when the loss value is smaller than a preset loss threshold value, stopping training, and outputting the deep learning model to obtain the classification model.
Optionally, the screening all the classification categories according to the classification probability value to obtain a classification result includes:
judging whether the classification probability value is larger than a preset threshold value or not;
if the classification probability value is larger than a preset threshold value, selecting the classification probability value larger than the preset threshold value to obtain a classification probability value set;
selecting the maximum value in the classification probability value set to obtain a target probability value;
taking the classification category corresponding to the target probability value as the classification result;
and if the classification probability value is not greater than a preset threshold value, transmitting all the classification probability values to preset terminal equipment.
In order to solve the above problems, the present invention also provides an information classification apparatus, the apparatus comprising:
the data dividing module is used for acquiring a user history information set and dividing the user history information set into a first history information set and a second history information set by utilizing a preset field;
the model training module is used for training the pre-constructed deep learning model by utilizing the first historical information set to obtain a first classification model; performing category label marking on each user history information in the second history information set by using the first classification model; training the deep learning model by using the second historical information set and the first historical information set marked by the labels to obtain the second classification model;
the information classification module is used for classifying the user information to be evaluated by utilizing the second classification model when receiving the user information to be classified, so as to obtain classification probability values of a plurality of classification categories; and screening all the classification categories according to the classification probability value to obtain classification results.
In order to solve the above-mentioned problems, the present invention also provides an electronic apparatus including:
a memory storing at least one computer program; and
And a processor executing the computer program stored in the memory to implement the information classification method.
In order to solve the above-described problems, the present invention also provides a computer-readable storage medium having stored therein at least one computer program that is executed by a processor in an electronic device to implement the above-described information classification method.
In the embodiment of the invention, the first classification model is utilized to label each user history information in the second history information set, the second history information set marked by labels and the first history information set are utilized to train the deep learning model, the second classification model is obtained, the data without labels are marked by the model, and then the marked data are combined with the original data to retrain the model, so that the performance of the model is improved, and the accuracy of information classification is improved; therefore, the information classification method, the information classification device, the electronic equipment and the readable storage medium provided by the embodiment of the invention improve the accuracy of information classification.
Drawings
Fig. 1 is a flow chart of an information classification method according to an embodiment of the invention;
FIG. 2 is a schematic block diagram of an information classification device according to an embodiment of the invention;
fig. 3 is a schematic diagram of an internal structure of an electronic device for implementing an information classification method according to an embodiment of the present invention;
the achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
The embodiment of the invention provides an information classification method. The execution subject of the information classification method includes, but is not limited to, at least one of a server, a terminal, and the like, which can be configured to execute the method provided in the embodiments of the present application. In other words, the information classification method may be performed by software or hardware installed in a terminal device or a server device, and the software may be a blockchain platform. The service end includes but is not limited to: the server can be an independent server, or can be a cloud server for providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDNs), basic cloud computing services such as big data and artificial intelligent platforms, and the like.
Referring to fig. 1, a flowchart of an information classification method according to an embodiment of the present invention is shown, where in the embodiment of the present invention, the information classification method includes:
s1, acquiring a user history information set, and dividing the user history information set into a first history information set and a second history information set by utilizing a preset field;
in detail, in the embodiment of the present invention, the user history information set is a set of different user history information, and the user history information includes: user age, user type, user gender, region to which the user belongs, whether to transfer introduction, whether to WeChat the user, whether WeChat pays attention to public numbers, whether to take one person for multiple vehicles, whether to download vehicle owner software, last month of vehicle owner activity days, last two months of vehicle owner activity days, driving license type, license plate type, vehicle body color, vehicle type marketing time, vehicle seat number, displacement, whether to retrofit, whether to new energy vehicle, vehicle value, vehicle system, accumulated number of times of risk in the current year, number of times of risk in the past, historical number of risk, production resource type, maximum number of continuous years, total number of insurance in the department, vehicle risk order cost, whether to take insurance business, user category and other user characteristics in the user history information are objective characteristics, but the user category characteristics are subjective characteristics, and the user category comprises: the high-quality dangerous user, the high-quality continuous user, the high-quality premium user and the common user need the staff with rich experience to divide and mark, and the user history information does not have the corresponding user category characteristics, so the user history information set needs to be divided.
In detail, in the embodiment of the present invention, the dividing the user history information set into the first history information set and the second history information set by using the preset field includes: selecting all user history information with values of preset fields not being null values in the user history information set, and taking the values of the preset fields as class labels of corresponding user history information to obtain the first history information set; selecting all user history information with null values of preset fields in the user history information set to obtain the second history information set; wherein the preset field is a user category field.
For example: the value corresponding to the user category field in the user history information is "normal user", and then the category label corresponding to the user history information is "normal user".
Optionally, in the embodiment of the present invention, the user history information may be stored in a blockchain node, so as to improve the data taking efficiency according to the high throughput characteristic of the blockchain node.
S2, training a pre-constructed deep learning model by using the first historical information set to obtain a first classification model;
in detail, the deep learning model in the embodiment of the invention can be an artificial intelligent model, and a structure of a plurality of full-connection layers is adopted during the deep learning construction, so that the data expression capability of a network is enhanced, the complexity of the network is improved, and the accuracy of service distribution by using the network in the follow-up process is improved.
In an embodiment of the present invention, before the training of the pre-constructed deep learning model by using the first historical information set, the method further includes:
acquiring a deep neural network frame;
constructing a feature input layer in the deep neural network framework;
constructing a weight initialization layer after the feature input layer;
establishing a plurality of full connection layers after the characteristic input layer;
constructing batch standardization layers and discarding layers among the plurality of full connection layers;
and constructing a decision output layer after the plurality of full-connection layers to obtain the deep learning model.
In detail, the deep neural network frame can be preset by a user, and functions corresponding to different network levels are written in the deep neural network frame through java, python and other computer languages, so that the construction of a feature input layer, a weight initialization layer, a full connection layer, a batch standardization layer, a discarding layer and a decision output layer is realized.
Further, when training a pre-built deep learning model by using the first historical information set, performing feature division on the first historical information set by using the input layer to obtain different initial data features, performing model input, setting different weights for the input initial data features by the weight initialization layer, performing feature extraction on the initial data features with the weights by the full connection layer to obtain data features, selecting the data features extracted by the last full connection layer by using the decision output layer, and calculating by using a preset activation function to obtain prediction probabilities of different categories, wherein the batch standardization layer set is used for performing parameter adjustment on the full connection layer by the discarding layer.
Specifically, the feature input layer is used for dividing input data.
The full connection layer is used for expressing and analyzing the data characteristics input by the characteristic input layer according to preset weights so as to better display the hidden relation among the characteristics and further obtain training results of training data; the structure of the multi-layer full-connection layer is beneficial to increasing the complexity of the network so as to improve the accuracy of training results output by the network, and the multi-layer full-connection layer comprises an output layer of a model and is used for outputting the results of model analysis.
For example, when training the deep learning by using the first historical information set, the full-connection layer is used for expressing and analyzing the data features of the first historical information set, and the decision output layer after the plurality of full-connection layers is used for selecting the data features extracted and expressed by the last full-connection layer to calculate by using a preset activation function, so that the prediction probabilities of different categories are obtained.
The batch standardization layer is used for standardizing the data characteristics expressed by the full connection layer so as to solve the problem of gradient disappearance in the network training process, and can carry out weight adjustment on the full connection layer according to the data characteristics expressed by the preset weight so as to optimize the gradient flow of the network.
The discarding layer is a Dropout layer, and can temporarily discard the data features expressed by the full-connection layer according to preset probability parameters, so as to prevent the network from being in a fitting state when the training data are less. In one embodiment of the present invention, the deep learning includes an 8-layer network structure, the first layer is a feature input layer, the second layer is a weight initialization layer, the third layer is a full-connection layer including 64 neurons, the fourth layer is a batch normalization layer, the fifth layer is a discarding layer, the sixth layer is a full-connection layer including 32 neurons, the seventh layer is a full-connection layer including 16 neurons, the seventh layer is a full-connection layer (output layer) including 1 neuron, wherein the third layer and the sixth layer adopt a Relu function as an activation function, the seventh layer adopts a Sigmoid function as an activation function, and a probability parameter of the discarding layer is 0.3.
The output node of the decision output layer is set according to the category of each user history information category label in the first history information set, for example: the categories of each of the user history information category labels in the first set of history information are common: the decision output layer can be set to have 4 output nodes in total, wherein the first output node corresponds to the high-quality insurance user category, the second output node corresponds to the high-quality insurance user category, the third node corresponds to the high-quality premium user category, and the fourth node corresponds to the common user category.
In the embodiment of the present invention, training a pre-constructed deep learning model by using the first historical information set to obtain a first classification model includes:
step I, training the deep learning model by utilizing each user history information in the first history information set, and extracting an output value of a decision output layer in the deep learning model to obtain a classification predicted value;
step II, determining a classification true value according to each user history information category label in the first history information set;
step III, calculating a preset loss function according to the classification true value and the classification predicted value to obtain a loss value;
optionally, in an embodiment of the present invention, the loss function is a cross entropy loss function.
Step IV, when the loss value is larger than or equal to a preset loss threshold value, adjusting model parameters of the deep learning model according to the loss value, and returning to the step I; and when the loss value is smaller than a preset loss threshold value, stopping training, and outputting the deep learning model to obtain the classification model.
In one embodiment of the present invention, when the model parameters of the deep learning model are adjusted according to the loss value, a preset optimization algorithm may be used to adjust the parameters of the countermeasure generation network, where the preset optimization algorithm includes, but is not limited to: a batch gradient descent algorithm, a small batch gradient descent algorithm, and a random gradient descent algorithm.
For example, the current parameters of the deep learning model are input into the optimization algorithm, the input current parameters are optimally calculated by the optimization algorithm to obtain optimized parameters, and the optimized parameters are used for updating the current parameters of the countermeasure generation network to realize the adjustment of the parameters in the countermeasure generation network.
In an embodiment of the present invention, the adjusting the deep learning parameter according to the loss value includes:
calculating the update gradient of the parameters in the deep learning model by using a preset optimization algorithm according to the loss value;
and updating parameters of the deep learning model according to the updating gradient.
When the current parameters are updated according to the update gradient, the parameters in the deep learning model and the update gradient can be utilized to perform preset arithmetic operation.
For example, the current parameter is 10, the parameter in the deep learning is calculated by using a preset optimization algorithm, the update gradient of the parameter is 0.1, and the parameter can be updated to 10 (1+0.1) =11 according to the update gradient 0.1;
alternatively, the parameter may be updated to 10+0.1=10.1 according to the update gradient 0.1.
S3, marking the category label of each user history information in the second history information set by using the first classification model;
in detail, in the embodiment of the present invention, each of the history information in the second set of history information is input into the first classification model to obtain a class probability value of a plurality of classes; because the first classification model is trained by using a first historical information set, the category is consistent with the user category label corresponding to the first historical information set, and the method comprises the following steps: the high-quality insurance users, the high-quality renewal users, the high-quality premium users and the common users are classified into 4 categories.
Further, in the embodiment of the invention, a category probability value larger than a preset threshold value in all category probability values of each user history information is selected to obtain a category probability value set; judging the number of maximum values in the category probability value set, if the number of the maximum values is not equal to 1, deleting the user history information corresponding to the category probability value in the second history information set, and if the data of the maximum values is not equal to 1, indicating that the category characteristics of the corresponding user history information are not obvious, and helping the model to learn the corresponding category characteristics, so that the corresponding user history information is deleted; and if the number of the maximum values is equal to 1, determining the category corresponding to the maximum value in the category probability value set as the corresponding category label of the user history information.
S4, training the deep learning model by using the second historical information set and the first historical information set marked by the labels to obtain the second classification model;
in detail, in the embodiment of the present invention, the second history information set marked by the tag is combined with the first history information set to obtain an updated first history information set, and the deep learning model is trained by using the updated first history information set, where a specific training process is consistent with S3 and is not described herein in detail.
S5, when receiving the user information to be classified, classifying the user information to be evaluated by using the second classification model to obtain classification probability values of a plurality of classification categories;
in detail, in the embodiment of the present invention, the user information to be classified is unclassified user data having the same type of content as the history information in the second set of history information.
Further, in the embodiment of the present invention, the user information to be classified is input into the second classification model, so as to obtain classification probability values of multiple classification categories.
In detail, in the embodiment of the present invention, the second classification model is obtained through training the deep learning model, and the structure of the second classification model is the same as that of the deep learning model, so in the embodiment of the present invention, the user information to be classified is input into the second classification model, the output value of each output node of the decision output layer in the second classification model is extracted, and the classification probability value of the corresponding classification class is obtained, and when the deep learning model is constructed, the classification class corresponding to each output node is set in advance according to the user class label corresponding to the first historical information set, and therefore, according to the corresponding relation between the set output node and the classification class, the output value of the output node is used as the classification probability value of the corresponding classification class.
Such as: and S2, inputting the user information to be classified into the second classification model if the class corresponding to the first output node set in the deep learning model decision output layer is a high-quality risk user class, and extracting the output value of the first output node in the decision output layer in the second classification model as a classification probability value corresponding to the high-quality risk user class.
And S6, screening all the classification categories according to the classification probability value to obtain classification results.
In detail, in the embodiment of the invention, whether the classification probability value is larger than a preset threshold value is judged; if the classification probability value is larger than a preset threshold value, selecting the classification probability value larger than the preset threshold value to obtain a classification probability value set; selecting the maximum value in the classification probability value set to obtain a target probability value; and taking the preset category corresponding to the target probability value as the classification result, and if the classification probability value is not greater than a preset threshold value, transmitting all the classification probability values to preset terminal equipment, wherein the terminal equipment is terminal equipment of service personnel and comprises: cell phones, computers, tablets, etc.
Specifically, in the embodiment of the present invention, when the maximum value in the classification probability value set is selected, there may be a plurality of maximum values, so in the embodiment of the present invention, when the maximum value in the classification probability value set exists, one maximum value in the classification probability value set is arbitrarily selected, and the target probability value is obtained.
As shown in fig. 2, a functional block diagram of the information classification apparatus according to the present invention is shown.
The information classification apparatus 100 of the present invention may be installed in an electronic device. Depending on the functions implemented, the information classification means may comprise a data partitioning module 101, a model training module 102, an information classification module 103, which may also be referred to as a unit, refers to a series of computer program segments capable of being executed by a processor of an electronic device and of performing a fixed function, which are stored in a memory of the electronic device.
In the present embodiment, the functions concerning the respective modules/units are as follows:
the data dividing module 101 is configured to obtain a user history information set, and divide the user history information set into a first history information set and a second history information set by using a preset field;
the model training module 102 is configured to train a pre-constructed deep learning model by using the first historical information set to obtain a first classification model; performing category label marking on each user history information in the second history information set by using the first classification model; training the deep learning model by using the second historical information set and the first historical information set marked by the labels to obtain the second classification model;
the information classification module 103 is configured to, when receiving user information to be classified, classify the user information to be evaluated using the second classification model, and obtain classification probability values of a plurality of classification categories; and screening all the classification categories according to the classification probability value to obtain classification results.
In detail, each module in the information classification device 100 in the embodiment of the present invention adopts the same technical means as the information classification method described in fig. 1 and can produce the same technical effects when in use, and will not be described again here.
Fig. 2 is a schematic structural diagram of an electronic device for implementing the information classification method according to the present invention.
The electronic device may comprise a processor 10, a memory 11, a communication bus 12 and a communication interface 13, and may further comprise a computer program, such as an information classification program, stored in the memory 11 and executable on the processor 10.
The memory 11 includes at least one type of readable storage medium, including flash memory, a mobile hard disk, a multimedia card, a card memory (e.g., SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. The memory 11 may in some embodiments be an internal storage unit of the electronic device, such as a mobile hard disk of the electronic device. The memory 11 may in other embodiments also be an external storage device of the electronic device, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) or the like, which are provided on the electronic device. Further, the memory 11 may also include both an internal storage unit and an external storage device of the electronic device. The memory 11 may be used not only for storing application software installed in an electronic device and various types of data, such as codes of information classification programs, but also for temporarily storing data that has been output or is to be output.
The processor 10 may be comprised of integrated circuits in some embodiments, for example, a single packaged integrated circuit, or may be comprised of multiple integrated circuits packaged with the same or different functions, including one or more central processing units (Central Processing unit, CPU), microprocessors, digital processing chips, graphics processors, combinations of various control chips, and the like. The processor 10 is a Control Unit (Control Unit) of the electronic device, connects various components of the entire electronic device using various interfaces and lines, and executes various functions of the electronic device and processes data by running or executing programs or modules (e.g., information classification programs, etc.) stored in the memory 11, and calling data stored in the memory 11.
The communication bus 12 may be a peripheral component interconnect standard (perIPheral component interconnect, PCI) bus, or an extended industry standard architecture (extended industry standard architecture, EISA) bus, among others. The bus may be classified as an address bus, a data bus, a control bus, etc. The communication bus 12 is arranged to enable a connection communication between the memory 11 and at least one processor 10 etc. For ease of illustration, the figures are shown with only one bold line, but not with only one bus or one type of bus.
Fig. 2 shows only an electronic device with components, and it will be understood by those skilled in the art that the structure shown in fig. 2 is not limiting of the electronic device and may include fewer or more components than shown, or may combine certain components, or a different arrangement of components.
For example, although not shown, the electronic device may further include a power source (such as a battery) for supplying power to the respective components, and preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that functions of charge management, discharge management, power consumption management, and the like are implemented through the power management device. The power supply may also include one or more of any of a direct current or alternating current power supply, recharging device, power failure detection circuit, power converter or inverter, power status indicator, etc. The electronic device may further include various sensors, bluetooth modules, wi-Fi modules, etc., which are not described herein.
Optionally, the communication interface 13 may comprise a wired interface and/or a wireless interface (e.g., WI-FI interface, bluetooth interface, etc.), typically used to establish a communication connection between the electronic device and other electronic devices.
Optionally, the communication interface 13 may further comprise a user interface, which may be a Display, an input unit, such as a Keyboard (Keyboard), or a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like. The display may also be referred to as a display screen or display unit, as appropriate, for displaying information processed in the electronic device and for displaying a visual user interface.
It should be understood that the embodiments described are for illustrative purposes only and are not limited to this configuration in the scope of the patent application.
The information classification program stored in the memory 11 in the electronic device is a combination of a plurality of computer programs, which when run in the processor 10, can implement:
acquiring a user history information set, and dividing the user history information set into a first history information set and a second history information set by utilizing a preset field;
training a pre-constructed deep learning model by using the first historical information set to obtain a first classification model;
performing category label marking on each user history information in the second history information set by using the first classification model;
training the deep learning model by using the second historical information set and the first historical information set marked by the labels to obtain the second classification model;
when receiving user information to be classified, classifying the user information to be evaluated by using the second classification model to obtain classification probability values of a plurality of classification categories;
and screening all the classification categories according to the classification probability value to obtain classification results.
In particular, the specific implementation method of the processor 10 on the computer program may refer to the description of the relevant steps in the corresponding embodiment of fig. 1, which is not repeated herein.
Further, the electronic device integrated modules/units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. The computer readable medium may be non-volatile or volatile. The computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM).
Embodiments of the present invention may also provide a computer readable storage medium storing a computer program which, when executed by a processor of an electronic device, may implement:
acquiring a user history information set, and dividing the user history information set into a first history information set and a second history information set by utilizing a preset field;
training a pre-constructed deep learning model by using the first historical information set to obtain a first classification model;
performing category label marking on each user history information in the second history information set by using the first classification model;
training the deep learning model by using the second historical information set and the first historical information set marked by the labels to obtain the second classification model;
when receiving user information to be classified, classifying the user information to be evaluated by using the second classification model to obtain classification probability values of a plurality of classification categories;
and screening all the classification categories according to the classification probability value to obtain classification results.
Further, the computer-usable storage medium may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function, and the like; the storage data area may store data created from the use of blockchain nodes, and the like.
In the several embodiments provided in the present invention, it should be understood that the disclosed apparatus, device and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be other manners of division when actually implemented.
The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
The embodiment of the application can acquire and process the related data based on the artificial intelligence technology. Among these, artificial intelligence (Artificial Intelligence, AI) is the theory, method, technique and application system that uses a digital computer or a digital computer-controlled machine to simulate, extend and extend human intelligence, sense the environment, acquire knowledge and use knowledge to obtain optimal results.
In addition, each functional module in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units can be realized in a form of hardware or a form of hardware and a form of software functional modules.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof.
The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
The blockchain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, encryption algorithm and the like. The Blockchain (Blockchain), which is essentially a decentralised database, is a string of data blocks that are generated by cryptographic means in association, each data block containing a batch of information of network transactions for verifying the validity of the information (anti-counterfeiting) and generating the next block. The blockchain may include a blockchain underlying platform, a platform product services layer, an application services layer, and the like.
Furthermore, it is evident that the word "comprising" does not exclude other elements or steps, and that the singular does not exclude a plurality. A plurality of units or means recited in the system claims can also be implemented by means of software or hardware by means of one unit or means. The terms second, etc. are used to denote a name, but not any particular order.
Finally, it should be noted that the above-mentioned embodiments are merely for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention.
Claims (6)
1. An information classification method, the method comprising:
acquiring a user history information set, and dividing the user history information set into a first history information set and a second history information set by utilizing a preset field;
training a pre-constructed deep learning model by using the first historical information set to obtain a first classification model;
performing category label marking on each user history information in the second history information set by using the first classification model;
training the deep learning model by using the second historical information set and the first historical information set marked by the labels to obtain a second classification model;
when receiving user information to be classified, classifying the user information to be classified by using the second classification model to obtain classification probability values of a plurality of classification categories;
screening all the classification categories according to the classification probability value to obtain classification results;
the step of dividing the user history information set into a first history information set and a second history information set by using a preset field includes: selecting all user history information with values of preset fields not being null values in the user history information set, and taking the values of the preset fields as class labels of corresponding user history information to obtain the first history information set; selecting all user history information with null values of preset fields in the user history information set to obtain the second history information set;
the performing category label marking on each user history information in the second history information set by using the first classification model includes: inputting each user history information in the second history information set into the first classification model to obtain class probability values of a plurality of classes; selecting a class probability value which is larger than a preset threshold value from all class probability values of each user history information to obtain a class probability value set; judging the number of maximum values in the class probability value set; if the number of the maximum values is not equal to 1, deleting the user history information corresponding to the category probability value set in the second history information set; if the number of the maximum values is not equal to 1, determining the category corresponding to the maximum value in the category probability value set as the category label of the corresponding user history information;
the training of the deep learning model by using the second historical information set and the first historical information set of the label mark to obtain the second classification model comprises the following steps: combining the second historical information set marked by the label with the first historical information set to obtain an updated first historical information set; training the deep learning model by using the updated first historical information set to obtain the second classification model;
screening all the classification categories according to the classification probability value to obtain classification results, wherein the method comprises the following steps: judging whether the classification probability value is larger than a preset threshold value or not; if the classification probability value is larger than a preset threshold value, selecting the classification probability value larger than the preset threshold value to obtain a classification probability value set; selecting the maximum value in the classification probability value set to obtain a target probability value; taking the classification category corresponding to the target probability value as the classification result; and if the classification probability value is not greater than a preset threshold value, transmitting all the classification probability values to preset terminal equipment.
2. The information classification method of claim 1, wherein prior to training the pre-constructed deep learning model using the first set of historical information, the method further comprises:
acquiring a deep neural network frame;
constructing a feature input layer in the deep neural network framework;
constructing a weight initialization layer after the feature input layer;
establishing a plurality of full connection layers after the characteristic input layer;
constructing batch standardization layers and discarding layers among the plurality of full connection layers;
and constructing a decision output layer after the plurality of full-connection layers to obtain the deep learning model.
3. The information classification method of claim 1, wherein prior to training the pre-constructed deep learning model using the first set of historical information, the method further comprises:
selecting user history information in the first history information set, inputting the user history information into the deep learning model, and extracting an output value of a decision output layer in the deep learning model to obtain a classification predicted value;
determining a classification authenticity value according to each user history information category label in the first history information set;
calculating a preset loss function according to the classification reality value and the classification predicted value to obtain a loss value;
when the loss value is greater than or equal to a preset loss threshold value, adjusting model parameters of the deep learning model according to the loss value, and returning the user history information in the selected first history information set to input the deep learning model;
and when the loss value is smaller than a preset loss threshold value, stopping training, and outputting the deep learning model to obtain the classification model.
4. An information classification apparatus for implementing the information classification method according to any one of claims 1 to 3, characterized in that the information classification apparatus comprises:
the data dividing module is used for acquiring a user history information set and dividing the user history information set into a first history information set and a second history information set by utilizing a preset field;
the model training module is used for training the pre-constructed deep learning model by utilizing the first historical information set to obtain a first classification model; performing category label marking on each user history information in the second history information set by using the first classification model; training the deep learning model by using the second historical information set and the first historical information set marked by the labels to obtain a second classification model;
the information classification module is used for classifying the user information to be classified by utilizing the second classification model when receiving the user information to be classified, so as to obtain classification probability values of a plurality of classification categories; and screening all the classification categories according to the classification probability value to obtain classification results.
5. An electronic device, the electronic device comprising:
at least one processor; the method comprises the steps of,
a memory communicatively coupled to the at least one processor;
wherein the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the information classification method of any one of claims 1 to 3.
6. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the information classification method according to any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111011806.7A CN113626606B (en) | 2021-08-31 | 2021-08-31 | Information classification method, device, electronic equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111011806.7A CN113626606B (en) | 2021-08-31 | 2021-08-31 | Information classification method, device, electronic equipment and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113626606A CN113626606A (en) | 2021-11-09 |
| CN113626606B true CN113626606B (en) | 2024-02-13 |
Family
ID=78388488
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111011806.7A Active CN113626606B (en) | 2021-08-31 | 2021-08-31 | Information classification method, device, electronic equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113626606B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114398514B (en) * | 2021-12-24 | 2022-11-22 | 北京达佳互联信息技术有限公司 | Video display method and device and electronic equipment |
| CN114547448B (en) * | 2022-02-17 | 2023-09-01 | 北京百度网讯科技有限公司 | Data processing method, model training method, device, equipment, storage medium and program |
| CN114936942B (en) * | 2022-07-21 | 2022-11-01 | 深圳市绽放工场科技有限公司 | Computer network data processing and analyzing system and method for insurance users |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110990562A (en) * | 2019-10-29 | 2020-04-10 | 新智认知数字科技股份有限公司 | Alarm classification method and system |
| CN112860905A (en) * | 2021-04-08 | 2021-05-28 | 深圳壹账通智能科技有限公司 | Text information extraction method, device and equipment and readable storage medium |
| CN113240130A (en) * | 2020-06-22 | 2021-08-10 | 北京芯盾时代科技有限公司 | Data classification method and device, computer readable storage medium and electronic equipment |
| CN113268665A (en) * | 2021-05-24 | 2021-08-17 | 平安科技（深圳）有限公司 | Information recommendation method, device and equipment based on random forest and storage medium |
-
2021
- 2021-08-31 CN CN202111011806.7A patent/CN113626606B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110990562A (en) * | 2019-10-29 | 2020-04-10 | 新智认知数字科技股份有限公司 | Alarm classification method and system |
| CN113240130A (en) * | 2020-06-22 | 2021-08-10 | 北京芯盾时代科技有限公司 | Data classification method and device, computer readable storage medium and electronic equipment |
| CN112860905A (en) * | 2021-04-08 | 2021-05-28 | 深圳壹账通智能科技有限公司 | Text information extraction method, device and equipment and readable storage medium |
| CN113268665A (en) * | 2021-05-24 | 2021-08-17 | 平安科技（深圳）有限公司 | Information recommendation method, device and equipment based on random forest and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113626606A (en) | 2021-11-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113822494B (en) | Risk prediction method, device, equipment and storage medium | |
| CN113626606B (en) | Information classification method, device, electronic equipment and readable storage medium | |
| CN112541745B (en) | User behavior data analysis method and device, electronic equipment and readable storage medium | |
| CN112801718B (en) | User behavior prediction method, device, equipment and medium | |
| CN112115322B (en) | User grouping method, device, electronic equipment and storage medium | |
| CN113656690B (en) | Product recommendation method and device, electronic equipment and readable storage medium | |
| CN113688923A (en) | Intelligent order abnormity detection method and device, electronic equipment and storage medium | |
| CN114781832A (en) | Course recommendation method and device, electronic equipment and storage medium | |
| CN114612194A (en) | Product recommendation method and device, electronic equipment and storage medium | |
| CN114358854A (en) | Customer loss early warning method, device, equipment and storage medium | |
| CN115081538A (en) | Customer relationship identification method, device, equipment and medium based on machine learning | |
| CN114880449A (en) | Reply generation method and device of intelligent question answering, electronic equipment and storage medium | |
| CN112269875B (en) | Text classification method, device, electronic equipment and storage medium | |
| CN113658002A (en) | Decision tree-based transaction result generation method and device, electronic equipment and medium | |
| CN116843481A (en) | Knowledge graph analysis method, device, equipment and storage medium | |
| CN112651782A (en) | Behavior prediction method, device, equipment and medium based on zoom dot product attention | |
| CN116401606A (en) | Fraud identification method, device, equipment and medium | |
| CN113806540B (en) | Text labeling method, text labeling device, electronic equipment and storage medium | |
| CN113706019B (en) | Service capability analysis method, device, equipment and medium based on multidimensional data | |
| CN113780473B (en) | Depth model-based data processing method and device, electronic equipment and storage medium | |
| CN115221274A (en) | Text emotion classification method and device, electronic equipment and storage medium | |
| CN113344415A (en) | Deep neural network-based service distribution method, device, equipment and medium | |
| CN113657546A (en) | Information classification method and device, electronic equipment and readable storage medium | |
| CN113486646A (en) | Product report issuing method and device, electronic equipment and readable storage medium | |
| CN113902418A (en) | Project approval method and device, electronic equipment and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |