CN110348581B - User feature optimizing method, device, medium and electronic equipment in user feature group - Google Patents
User feature optimizing method, device, medium and electronic equipment in user feature group Download PDFInfo
- Publication number
- CN110348581B CN110348581B CN201910533729.8A CN201910533729A CN110348581B CN 110348581 B CN110348581 B CN 110348581B CN 201910533729 A CN201910533729 A CN 201910533729A CN 110348581 B CN110348581 B CN 110348581B
- Authority
- CN
- China
- Prior art keywords
- user
- user feature
- score
- combination
- combinations
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Image Analysis (AREA)
- Machine Translation (AREA)
Abstract
The disclosure relates to a method, a device, a medium and an electronic device for optimizing user characteristics in a user characteristic group, belonging to the technical field of machine learning application, wherein the method comprises the following steps: initializing and generating a plurality of user feature combinations; inputting each user characteristic combination and the optimizing target into a first machine learning model to obtain an evaluation score; acquiring a plurality of user feature combinations in a first score range and a plurality of user feature combinations in a second score range; acquiring user characteristic distinction of the user characteristic combination of each first score range and the user characteristic combination of the second score range; inputting the distinction and optimization targets of the user features into a second machine learning model to obtain a prediction optimization user feature combination; and obtaining a target optimization user feature combination. According to the method and the device, the user characteristic combination corresponding to the optimizing target is predicted according to the distinguishing characteristics of part of the user characteristic combinations through the machine learning model, so that the high efficiency and accuracy of obtaining the target user characteristic combination are guaranteed.
Description
Technical Field
The disclosure relates to the technical field of machine learning application, in particular to a user feature optimizing method, a device, a medium and electronic equipment in a user feature group.
Background
The user feature optimizing in the user feature group is to search the user feature combination with special purpose meeting the requirement from the user feature group with a large number of user features, namely to search the optimized user features from the user feature group. For example, a user feature combination satisfying the demand is found from a user feature group in order to obtain a group of user combinations that can satisfy the predetermined purpose.
At present, the user characteristic optimization in the user characteristic group is mainly carried out by crawling keywords from the user characteristic group, setting corresponding weight coefficients based on different keywords, and then calculating user characteristic combinations meeting requirements through the weight coefficients. In this way, in the process of optimizing the user characteristics, the distinction between the user characteristics in all the user characteristic groups cannot be considered, the obtained user characteristic combinations have the problem of unreasonable combination, and the process of obtaining the user characteristic combinations needs to count a large amount of data, so that the algorithm load is large, and the obtaining efficiency is low.
It should be noted that the information disclosed in the above background section is only for enhancing understanding of the background of the present disclosure and thus may include information that does not constitute prior art known to those of ordinary skill in the art.
Disclosure of Invention
The purpose of the present disclosure is to provide a user feature optimizing scheme in a user feature group, so as to accurately and efficiently predict a user feature combination corresponding to an optimizing target according to distinguishing features of part of user feature combinations at least to a certain extent through a machine learning model, thereby ensuring high efficiency and accuracy of obtaining the target user feature combination.
According to one aspect of the present disclosure, there is provided a user feature optimizing method in a user feature group, including:
initializing and generating a plurality of user feature combinations based on the target user feature group;
respectively inputting each user feature combination and optimizing target into a first machine learning model trained in advance to obtain an evaluation score of each user feature combination;
acquiring user characteristic combinations corresponding to evaluation scores in a plurality of first score ranges and user characteristic combinations corresponding to evaluation scores in a plurality of second score ranges, wherein the evaluation scores in the plurality of first score ranges are larger than a first preset threshold value, and the evaluation scores in the plurality of second score ranges are lower than a second preset threshold value;
Acquiring user characteristic differences of the user characteristic combinations of each first score range and the user characteristic combinations of each second score range, and obtaining a plurality of user characteristic differences;
inputting the multiple user feature distinction and optimization targets into a pre-trained second machine learning model to obtain a predictive optimization user feature combination of the target user feature group;
and acquiring target optimized user feature combinations from the target user feature groups based on the predicted optimized user feature combinations.
In an exemplary embodiment of the present disclosure, the initializing generating a plurality of user feature combinations based on the target user feature group includes:
sequentially and respectively acquiring a preset number of user features from the target user feature group, wherein the preset number is more than the preset number in the previous time when each acquisition is performed;
and sequentially and respectively arranging the acquired preset number of user features into user feature strings according to the arrangement rule of the target user feature groups to obtain a plurality of user feature combinations.
In an exemplary embodiment of the present disclosure, the initializing generating a plurality of user feature combinations based on the target user feature group includes:
Based on the target user feature group, a predetermined number of user feature combinations are randomly initialized.
In an exemplary embodiment of the present disclosure, after the obtaining the user feature combinations corresponding to the evaluation scores in the plurality of first score ranges and the user feature combinations corresponding to the evaluation scores in the plurality of second score ranges, wherein the evaluation scores in the plurality of first score ranges are greater than a first predetermined threshold, the evaluation scores in the plurality of second score ranges are lower than a second predetermined threshold, the method further includes:
if the first score range and the second score range do not acquire the corresponding user feature combinations, initializing and generating a preset group of a plurality of user feature combinations based on the target user feature group;
respectively inputting the user feature combinations into a pre-trained first machine learning model to obtain evaluation scores of each user feature combination;
acquiring user characteristic combinations corresponding to evaluation scores of a plurality of first score ranges larger than a first preset threshold value and user characteristic combinations corresponding to evaluation scores of a plurality of second score ranges lower than a second preset threshold value in all the evaluation scores
And when all the first score range and the second score range acquire the corresponding user feature combination, ending the acquisition.
In an exemplary embodiment of the present disclosure, the acquiring acquires a user feature difference of the user feature combination of each of the first score ranges and the user feature combination of each of the second score ranges, and obtains a plurality of user feature differences, including:
randomly acquiring a user characteristic combination from the user characteristic combinations in each first score range;
randomly acquiring a user characteristic combination from the user characteristic combinations in each second score range;
and acquiring the user characteristic difference of the randomly acquired one user characteristic combination of each first score range and the randomly acquired one user characteristic combination of each second score range, and obtaining a plurality of user characteristic differences.
In an exemplary embodiment of the disclosure, the obtaining the user feature differences of the user feature combinations of each of the first score ranges and the user feature combinations of each of the second score ranges, includes:
acquiring first user characteristic combination elements of the user characteristic combinations of each first score range;
Acquiring second user characteristic combination elements of the user characteristic combination of the second score range corresponding to each first score range;
and obtaining distinguishing features of each first user feature combination element and each second user feature combination element, and obtaining user feature distinction of the user feature combination of each first score range and the user feature combination of each second score range.
In one exemplary embodiment of the present disclosure, the training method of the first machine learning model:
collecting a user characteristic combination and an optimizing target sample set, wherein each sample in the sample set is calibrated with a corresponding evaluation score in advance;
respectively inputting the input data of each sample in the sample set into a machine learning model to obtain an evaluation score corresponding to each sample;
if the input data with the sample is input into the machine learning model, obtaining that the evaluation score corresponding to the sample is inconsistent with the evaluation score calibrated in advance for the sample, and adjusting the coefficient of the learning model until the evaluation score is consistent with the evaluation score calibrated in advance for the sample;
after the input data of all the samples are input into the machine learning model, the obtained evaluation score corresponding to each sample is consistent with the evaluation score calibrated in advance for each sample, and training is finished.
According to one aspect of the present disclosure, there is provided a user feature optimizing apparatus in a user feature group, including:
the initialization module is used for initializing and generating a plurality of user feature combinations based on the target user feature groups;
the evaluation module is used for inputting each user characteristic combination and the optimizing target into a first machine learning model trained in advance respectively to obtain an evaluation score of each user characteristic combination;
the first acquisition module is used for acquiring user characteristic combinations corresponding to the evaluation scores in a plurality of first score ranges and user characteristic combinations corresponding to the evaluation scores in a plurality of second score ranges, wherein the evaluation scores in the plurality of first score ranges are larger than a first preset threshold value, and the evaluation scores in the plurality of second score ranges are lower than a second preset threshold value;
the second acquisition module is used for acquiring the user characteristic differences of the user characteristic combination of each first score range and the user characteristic combination of each second score range to obtain a plurality of user characteristic differences;
the prediction module is used for inputting the multiple user feature distinction and optimization targets into a pre-trained second machine learning model to obtain a predicted and optimized user feature combination of the target user feature group;
And the optimized feature acquisition module is used for acquiring the target optimized user feature combination from the target user feature group based on the predicted optimized user feature combination.
According to an aspect of the present disclosure, there is provided a computer readable storage medium having stored thereon a user feature optimizing program in a user feature group, wherein the user feature optimizing program in the user feature group, when executed by a processor, implements the method of any one of the above.
According to an aspect of the present disclosure, there is provided an electronic apparatus, including:
a processor; and
the memory is used for storing a user characteristic optimizing program in the user characteristic group of the processor; wherein the processor is configured to perform the method of any of the above via execution of a user feature optimization program in the user feature group.
The present disclosure relates to a method and a device for optimizing user characteristics in a user characteristic group, firstly, initializing and generating a plurality of user characteristic combinations based on a target user characteristic group; thus, a few user feature combinations can be quickly obtained, and further, each user feature combination and optimization target are respectively input into a first machine learning model trained in advance to obtain an evaluation score of each user feature combination; the suitability of each user characteristic combination and the optimizing target can be accurately evaluated. Then, obtaining user characteristic combinations corresponding to the evaluation scores in a plurality of first score ranges and user characteristic combinations corresponding to the evaluation scores in a plurality of second score ranges, wherein the evaluation scores in the plurality of first score ranges are larger than a first preset threshold value, and the evaluation scores in the plurality of second score ranges are lower than a second preset threshold value; in this way, a plurality of user feature combinations can be accurately classified according to the evaluation scores, and further, a plurality of user feature differences are obtained by obtaining the user feature differences of the user feature combinations of each first score range and the user feature combinations of each second score range; the user characteristic distinction in a series range can be accurately obtained according to the comparison of the user characteristic combinations of different score categories, and the reliability of optimizing the user characteristic combination analysis of the distinguishing characteristics in the subsequent steps is effectively ensured. Then, inputting the multiple user feature distinction and optimization targets into a pre-trained second machine learning model to obtain a predicted and optimized user feature combination of the target user feature group; in this way, the optimal user feature combination can be predicted efficiently and accurately based on the user feature distinction of the series range. Furthermore, based on the predicted optimal user feature combination, the target optimal user feature combination can be efficiently and accurately obtained from the target user feature group. In this way, the user feature combination corresponding to the optimizing target is accurately and efficiently predicted according to the distinguishing features of part of the user feature combinations through the machine learning model, so that the high efficiency and accuracy of the acquisition of the target user feature combination are ensured.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description, serve to explain the principles of the disclosure. It will be apparent to those of ordinary skill in the art that the drawings in the following description are merely examples of the disclosure and that other drawings may be derived from them without undue effort.
Fig. 1 schematically shows a flow chart of a method for optimizing user characteristics in a user characteristic group.
Fig. 2 schematically shows an exemplary diagram of an application scenario of a user feature optimizing method in a user feature group.
Fig. 3 schematically shows a flow chart of a method of obtaining a user characteristic distinction.
Fig. 4 schematically shows a block diagram of a user profile optimizing apparatus in a user profile.
Fig. 5 schematically shows an example block diagram of an electronic device for implementing the above-described method of user feature optimization in a user feature group.
Fig. 6 schematically illustrates a computer readable storage medium for implementing the user feature optimizing method in the user feature group described above.
Detailed Description
Example embodiments will now be described more fully with reference to the accompanying drawings. However, the exemplary embodiments may be embodied in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the present disclosure. One skilled in the relevant art will recognize, however, that the aspects of the disclosure may be practiced without one or more of the specific details, or with other methods, components, devices, steps, etc. In other instances, well-known technical solutions have not been shown or described in detail to avoid obscuring aspects of the present disclosure.
Furthermore, the drawings are merely schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus a repetitive description thereof will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor devices and/or microcontroller devices.
In this exemplary embodiment, a method for optimizing a user feature in a user feature group is provided first, where the method for optimizing a user feature in a user feature group may be run on a server, or may be run on a server cluster or a cloud server, or the like, and of course, a person skilled in the art may also run the method of the present invention on other platforms according to a requirement, which is not limited in particular in this exemplary embodiment. Referring to fig. 1, the method for optimizing the user characteristics in the user characteristic group may include the following steps:
step S110, initializing and generating a plurality of user feature combinations based on the target user feature group;
step S120, inputting each user feature combination and optimizing target into a first machine learning model trained in advance respectively to obtain an evaluation score of each user feature combination;
step S130, obtaining user characteristic combinations corresponding to evaluation scores in a plurality of first score ranges and user characteristic combinations corresponding to evaluation scores in a plurality of second score ranges, wherein the evaluation scores in the plurality of first score ranges are larger than a first preset threshold value, and the evaluation scores in the plurality of second score ranges are lower than a second preset threshold value;
Step S140, obtaining the user characteristic differences of the user characteristic combination of each first score range and the user characteristic combination of each second score range, and obtaining a plurality of user characteristic differences;
step S150, inputting the multiple user feature distinction and optimization targets into a pre-trained second machine learning model to obtain a predicted and optimized user feature combination of the target user feature group;
step S160, obtaining a target optimized user feature combination from the target user feature group based on the predicted optimized user feature combination.
In the user feature optimizing method in the user feature group, firstly, a plurality of user feature combinations are generated in an initializing mode based on a target user feature group; thus, a few user feature combinations can be quickly obtained, and further, each user feature combination and optimization target are respectively input into a first machine learning model trained in advance to obtain an evaluation score of each user feature combination; the suitability of each user characteristic combination and the optimizing target can be accurately evaluated. Then, obtaining user characteristic combinations corresponding to the evaluation scores in a plurality of first score ranges and user characteristic combinations corresponding to the evaluation scores in a plurality of second score ranges, wherein the evaluation scores in the plurality of first score ranges are larger than a first preset threshold value, and the evaluation scores in the plurality of second score ranges are lower than a second preset threshold value; in this way, a plurality of user feature combinations can be accurately classified according to the evaluation scores, and further, a plurality of user feature differences are obtained by obtaining the user feature differences of the user feature combinations of each first score range and the user feature combinations of each second score range; the user characteristic distinction in a series range can be accurately obtained according to the comparison of the user characteristic combinations of different score categories, and the reliability of optimizing the user characteristic combination analysis of the distinguishing characteristics in the subsequent steps is effectively ensured. Then, inputting the multiple user feature distinction and optimization targets into a pre-trained second machine learning model to obtain a predicted and optimized user feature combination of the target user feature group; in this way, the optimal user feature combination can be predicted efficiently and accurately based on the user feature distinction of the series range. Furthermore, based on the predicted optimal user feature combination, the target optimal user feature combination can be efficiently and accurately obtained from the target user feature group. In this way, the user feature combination corresponding to the optimizing target is accurately and efficiently predicted according to the distinguishing features of part of the user feature combinations through the machine learning model, so that the high efficiency and accuracy of the acquisition of the target user feature combination are ensured.
Next, each step in the user feature optimizing method in the user feature group in the present exemplary embodiment will be explained and described in detail with reference to the accompanying drawings.
In step S110, a plurality of user feature combinations are initially generated based on the target user feature group.
In the embodiment of the present example, referring to fig. 2, a server 201 acquires a target user feature group from a platform server 202, and then generates a plurality of user feature combinations based on the target user feature group by initialization. In this way, the server 201 can perform the combined feature analysis of the user feature combination according to the plurality of user feature combinations generated by initialization in the subsequent step, and thus the optimization target user combined feature can be obtained. It will be appreciated that, depending on the requirements, multiple user feature combinations may be generated directly by the server 202 based on the target user feature groups. And in the subsequent step, carrying out user characteristic combination characteristic analysis according to the plurality of user characteristic combinations generated by initialization, and further obtaining the optimization target user characteristics. The server 201 and the server 202 may be any devices with processing capability, such as a computer, a microprocessor, a cloud server, etc., which are not limited herein.
The target user feature group is a user feature group including a large number of user features, for example, a large number of user features on a shopping platform or a communication platform. Based on the target user feature group, initializing and generating a plurality of user feature combinations, namely, acquiring a plurality of user features from the target user feature group and then combining the user feature combinations randomly or according to a preset sequence, and further acquiring the plurality of user feature combinations, wherein a few user feature combinations in all possible user feature combinations can be quickly acquired for evaluation in the subsequent step. The number of the obtained user feature combinations may be that the target user feature groups are arranged and combined according to the number of the user features, so as to obtain a predetermined proportion of the user feature combinations, and the predetermined proportion may be, for example, 0.001%, 0.002% or the like according to the requirement.
In one implementation manner of this example, the initializing to generate a plurality of user feature combinations based on the target user feature group includes:
sequentially and respectively acquiring a preset number of user features from the target user feature group, wherein the preset number is more than the preset number in the previous time when each acquisition is performed;
And sequentially and respectively arranging the acquired preset number of user features into user feature strings according to the arrangement rule of the target user feature groups to obtain a plurality of user feature combinations.
And sequentially and respectively acquiring a predetermined number of user features from the target user feature group, wherein the predetermined number is greater than the predetermined number in each acquisition, namely, for example, 100, 110 and 200 user features are sequentially acquired from the user feature group with 10000 user features, and the number of the user features is greater than 10 in each acquisition. And then, sequentially and respectively arranging the acquired preset number of user features into user feature strings according to the arrangement rule of the target user feature groups to obtain a plurality of user feature combinations. Therefore, when the target user feature group has own user feature arrangement rules, the acquired user feature combination is ensured to carry the user feature group features, and the accuracy of subsequent step analysis is ensured. The arrangement rule may be, for example, arrangement in a predetermined order or arrangement in a predetermined table form, or the like.
In one implementation manner of this example, the initializing to generate a plurality of user feature combinations based on the target user feature group includes:
Based on the target user feature group, a predetermined number of user feature combinations are randomly initialized.
Therefore, the randomness of the acquisition of the user feature combinations can be ensured by generating the preset number of the user feature combinations through random initialization, and the efficiency of the acquisition of the user feature combinations is ensured when the user feature groups are not arranged in sequence.
In step S120, each of the user feature combinations and the optimization targets is input into a first machine learning model trained in advance, so as to obtain an evaluation score of each of the user feature combinations.
In this exemplary embodiment, each user feature combination and the optimizing target are respectively input into a pre-trained machine learning model, and the suitability of each user feature combination and the optimizing target, that is, the evaluation score of each user feature combination, can be evaluated by the machine learning model according to the optimizing target. Wherein the optimization objective is the use of user feature combinations, e.g. whether certain user feature combinations can be adapted to the user combination requirements for a certain item.
In one embodiment of the present example, the training method of the first machine learning model includes:
collecting a user characteristic combination and an optimizing target sample set, wherein each sample in the sample set is calibrated with a corresponding evaluation score in advance;
Respectively inputting the input data of each sample in the sample set into a machine learning model to obtain an evaluation score corresponding to each sample;
if the input data with the sample is input into the machine learning model, obtaining that the evaluation score corresponding to the sample is inconsistent with the evaluation score calibrated in advance for the sample, and adjusting the coefficient of the learning model until the evaluation score is consistent with the evaluation score calibrated in advance for the sample;
after the input data of all the samples are input into the machine learning model, the obtained evaluation score corresponding to each sample is consistent with the evaluation score calibrated in advance for each sample, and training is finished.
The user feature combination and optimizing target sample is the user feature combination and optimizing target sample which historically corresponds to a certain user feature group. The user characteristic combination and the optimizing target sample set are collected to be used as input of a machine learning model, and each sample is calibrated with corresponding evaluation scores by an expert in advance to be used as output of the machine learning model. Then, after the input data of all the samples are input into the machine learning model, the obtained evaluation score corresponding to each sample is consistent with the evaluation score calibrated in advance for each sample, training is finished, a trained first learning model is obtained, and training accuracy can be effectively ensured.
In step S130, a user feature combination corresponding to the evaluation score in a plurality of first score ranges and a user feature combination corresponding to the evaluation score in a plurality of second score ranges are obtained, wherein the evaluation score in the plurality of first score ranges is greater than a first predetermined threshold and the evaluation score in the plurality of second score ranges is lower than a second predetermined threshold.
In the embodiment of the present example, the first predetermined threshold is a threshold value at which the set judgment evaluation score is high, and the second predetermined threshold is a threshold value at which the set judgment evaluation score is low. The plurality of first score ranges greater than the first predetermined threshold are, for example, the plurality of first score ranges of 60-65, 65-70, etc. when the first predetermined threshold is 60, and the plurality of second score ranges lower than the second predetermined threshold are, for example, the plurality of second score ranges of 50-45, 45-40, etc. when the second predetermined threshold is 50. In this way, the user characteristic combinations corresponding to the evaluation scores of a plurality of first score ranges larger than a first preset threshold and the user characteristic combinations corresponding to the evaluation scores of a plurality of second score ranges lower than a second preset threshold are obtained, and the user characteristic combinations of different levels corresponding to the optimizing targets can be accurately classified according to the evaluation scores. In one example, the first predetermined threshold is equal to the second predetermined threshold as desired; in another example, the first predetermined threshold is greater than the second predetermined threshold, and the user-combined feature of the fractional range between the first predetermined threshold and the second predetermined threshold is discarded.
In one embodiment of the present example, after the obtaining the user feature combinations corresponding to the evaluation scores in the plurality of first score ranges and the user feature combinations corresponding to the evaluation scores in the plurality of second score ranges, wherein the evaluation scores in the plurality of first score ranges are greater than a first predetermined threshold, the evaluation scores in the plurality of second score ranges are lower than a second predetermined threshold, the method further includes:
if the first score range and the second score range do not acquire the corresponding user feature combinations, initializing and generating a preset group of a plurality of user feature combinations based on the target user feature group;
respectively inputting the user feature combinations into a pre-trained first machine learning model to obtain evaluation scores of each user feature combination;
acquiring user characteristic combinations corresponding to evaluation scores of a plurality of first score ranges larger than a first preset threshold value and user characteristic combinations corresponding to evaluation scores of a plurality of second score ranges lower than a second preset threshold value in all the evaluation scores
And when all the first score range and the second score range acquire the corresponding user feature combination, ending the acquisition.
In this way, if there are the first score ranges and the second score ranges for which no corresponding user feature combination is obtained, for example, when a plurality of first score ranges greater than a first predetermined threshold are, for example, 60, a plurality of first score ranges of 60-65, 65-70, etc., wherein a range of 60-65 does not obtain a user feature combination. And initializing and generating a preset group of a plurality of user feature combinations based on the target user feature group, and re-scoring and classifying until the corresponding user feature combinations are acquired in all the first score range and the second score range, and ending the acquisition. Corresponding user characteristic combinations of each first fraction range and each second fraction range can be ensured, and reliability of analysis in subsequent steps is ensured.
In step S140, a user feature difference between the user feature combination of each first score range and the user feature combination of each second score range is obtained, so as to obtain a plurality of user feature differences.
In this exemplary embodiment, each first score range corresponds to one of the second score ranges, that is, one second score range associated in advance with each first score range. For example, the plurality of first score ranges greater than the first predetermined threshold are, for example, a plurality of first score ranges of 60-65, 65-70, etc. when the first predetermined threshold is 60, and the plurality of second score ranges lower than the second predetermined threshold are, for example, a plurality of second score ranges of 50-45, 45-40, etc. when the second predetermined threshold is 50, the first score range of 60-65 corresponds to the second score range of 50-45, and the first score range of 65-70 corresponds to the second score range of 45-40. Thus, the user characteristic differences of the user characteristic combination of each first score range and the user characteristic combination of the second score range corresponding to each first score range are obtained, and a plurality of user characteristic differences of a series range can be obtained. The user feature distinction may be the difference between the mapping vector of the user feature combination with high score or the converted character string and the mapping vector of the user feature combination with low score or the converted character string. Wherein the mapping vector of user feature is obtained by vector mapping of user feature, for example, a high-scoring user feature is 50 men and 120 women, mapped to (1, 50, 21, 120, 22), wherein 1 is a gender dimension vector, 50, 120 is a number dimension, 21, 22 is a gender attribute value dimension, and if a low-scoring user feature vector is (1, 10, 21, 160, 22), their difference is expressed as (1, 50, 21, 120, 22) - (1, 10, 21, 160, 22).
Therefore, the distinguishing technical features exist based on the evaluation serialization of each user feature combination, and the accuracy of optimizing the user feature combination based on the user feature distinguishing prediction in the subsequent step is effectively ensured.
In one embodiment of the present example, the obtaining the user feature differences between the user feature combinations of each of the first score ranges and the user feature combinations of each of the second score ranges includes:
randomly acquiring a user characteristic combination from the user characteristic combinations in each first score range;
randomly acquiring a user characteristic combination from the user characteristic combinations in each second score range;
and acquiring the user characteristic difference of the randomly acquired one user characteristic combination of each first score range and the randomly acquired one user characteristic combination of each second score range, and obtaining a plurality of user characteristic differences.
There may be multiple combinations of user features in each first score range and each second score range. But one user feature combination in each score range may represent the user feature combination feature for that range. Further, by randomly acquiring a user feature combination from the user feature combinations of each of the first score ranges; randomly acquiring a user characteristic combination from the user characteristic combinations in each second score range; the user characteristic difference of one user characteristic combination of each first score range and one user characteristic combination of the second score range corresponding to each first score range can be obtained efficiently, and a plurality of user characteristic differences are obtained.
In one embodiment of the present example, referring to fig. 3, the obtaining the user feature differences between the user feature combinations of each of the first score ranges and the user feature combinations of each of the second score ranges includes:
step S310, obtaining first user characteristic combination elements of the user characteristic combinations of each first score range;
step S320, obtaining second user feature combination elements of the user feature combinations of the second score ranges corresponding to each first score range;
step S330, obtaining the distinguishing features of each first user feature combination element and each second user feature combination element, so as to obtain the user feature distinction between the user feature combination of each first score range and the user feature combination of each second score range.
The user feature combination element at least comprises the number of user features in the user feature combination, preset feature labels of the user features, attribute values of a plurality of attributes of the user features and the like, wherein the preset feature labels of the user features are the identity of a certain user, and the attribute values of the plurality of attributes of the user features are the male and female of the gender of the user, the size of the age and the like. Thus, by acquiring the first user feature combination element of the user feature combination of each first score range and acquiring the second user feature combination element of the user feature combination of the second score range corresponding to each first score range, and then acquiring the distinguishing feature of each first user feature combination element and the distinguishing feature of the second user feature combination element corresponding to the first user feature combination element, the distinguishing feature of each user feature combination can be accurately acquired. The accuracy of analysis in the subsequent steps is effectively ensured.
In step S150, the multiple user feature differences and the optimizing targets are input into a second machine learning model trained in advance, so as to obtain a predicted and optimized user feature combination of the target user feature group.
In the embodiment of the present example, the multiple user feature differences are user feature differences in a series range, which can accurately reflect the user feature combination score change and the user feature combination feature change, and further, the multiple user feature differences and the optimizing targets are input into the second machine learning model trained in advance, so that the optimizing targets can effectively and accurately obtain the predicted and optimized user feature combination adapted to the optimizing targets of the target user feature group based on the user feature difference analysis, thereby ensuring the efficiency and accuracy of obtaining the target user feature combination from the user feature group in the subsequent step.
In one embodiment of the present example, the training method of the second machine learning model includes:
collecting user characteristic distinction and optimizing target sample sets, wherein each sample in the sample sets is calibrated with corresponding optimized user characteristic combinations in advance;
respectively inputting the input data of each sample in the sample set into a machine learning model to obtain an optimized user characteristic combination corresponding to each sample;
If the input data with the sample is input into the machine learning model, obtaining that the optimized user characteristic combination corresponding to the sample is inconsistent with the optimized user characteristic combination calibrated in advance for the sample, and adjusting the coefficient of the learning model until the optimized user characteristic combination is consistent with the optimized user characteristic combination calibrated in advance for the sample;
after the input data of all the samples are input into the machine learning model, the obtained optimized user characteristic combination corresponding to each sample is consistent with the optimized user characteristic combination calibrated in advance for each sample, and training is finished.
The user characteristic distinguishing and optimizing target sample is the user characteristic distinguishing and optimizing target sample which corresponds to a certain user characteristic group in history. The user characteristic distinction and optimizing target sample set is collected as the input of the machine learning model, and each sample is calibrated by an expert in advance to obtain the corresponding optimizing user characteristic combination as the output of the machine learning model. And then, after the input data of all the samples are input into the machine learning model, the obtained optimized user characteristic combination corresponding to each sample is consistent with the optimized user characteristic combination calibrated in advance for each sample, and training is finished, so that a trained second learning model is obtained, and the training accuracy can be effectively ensured.
In step S160, a target optimized user feature combination is obtained from the target user feature group based on the predicted optimized user feature combination.
In the embodiment of the present example, based on the predicted optimized user feature combination, the target optimized user feature combination may be accurately obtained from the target user feature group by comparing the user feature labels or the user feature attribute names one by one.
Finally, the user characteristic combination corresponding to the optimizing target is accurately and efficiently predicted according to the distinguishing characteristics of part of the user characteristic combinations through the machine learning model, and further the high efficiency and accuracy of the target user characteristic combination acquisition are guaranteed.
The disclosure also provides a user feature optimizing device in the user feature group. Referring to fig. 4, the user feature optimizing apparatus in the user feature group may include a user feature initializing module 410, an evaluating module 420, a first obtaining module 430, a second obtaining module 440, a predicting module 450, and an optimized feature obtaining module 460. Wherein:
the initialization module 410 may be configured to initialize a plurality of user feature combinations based on the target user feature group;
the evaluation module 420 may be configured to input each of the user feature combinations and the optimization targets into a first machine learning model trained in advance, to obtain an evaluation score of each of the user feature combinations;
The first obtaining module 430 may be configured to obtain a user feature combination corresponding to an evaluation score in a plurality of first score ranges, where the evaluation score in the plurality of first score ranges is greater than a first predetermined threshold, and a user feature combination corresponding to an evaluation score in a plurality of second score ranges, where the evaluation score in the plurality of second score ranges is lower than a second predetermined threshold;
the second obtaining module 440 may be configured to obtain a user feature difference between the user feature combination of each of the first score ranges and the user feature combination of each of the second score ranges, to obtain a plurality of user feature differences;
the prediction module 450 may be configured to input the multiple user feature differences and the optimizing targets into a pre-trained second machine learning model, so as to obtain a predicted and optimized user feature combination of the target user feature group;
the optimized feature acquisition module 460 may be configured to acquire a target optimized user feature combination from the target user feature group based on the predicted optimized user feature combination.
The specific details of each module in the user feature optimizing device in the user feature group are described in detail in the user feature optimizing method in the corresponding user feature group, so that the details are not repeated here.
It should be noted that although in the above detailed description several modules or units of a device for action execution are mentioned, such a division is not mandatory. Indeed, the features and functionality of two or more modules or units described above may be embodied in one module or unit in accordance with embodiments of the present disclosure. Conversely, the features and functions of one module or unit described above may be further divided into a plurality of modules or units to be embodied.
Furthermore, although the steps of the methods in the present disclosure are depicted in a particular order in the drawings, this does not require or imply that the steps must be performed in that particular order or that all illustrated steps be performed in order to achieve desirable results. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step to perform, and/or one step decomposed into multiple steps to perform, etc.
From the above description of embodiments, those skilled in the art will readily appreciate that the example embodiments described herein may be implemented in software, or may be implemented in software in combination with the necessary hardware. Thus, the technical solution according to the embodiments of the present disclosure may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (may be a CD-ROM, a U-disk, a mobile hard disk, etc.) or on a network, including several instructions to cause a computing device (may be a personal computer, a server, a mobile terminal, or a network device, etc.) to perform the method according to the embodiments of the present disclosure.
In an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided.
Those skilled in the art will appreciate that the various aspects of the invention may be implemented as a system, method, or program product. Accordingly, aspects of the invention may be embodied in the following forms, namely: an entirely hardware embodiment, an entirely software embodiment (including firmware, micro-code, etc.) or an embodiment combining hardware and software aspects may be referred to herein as a "circuit," module "or" system.
An electronic device 500 according to such an embodiment of the invention is described below with reference to fig. 5. The electronic device 500 shown in fig. 5 is merely an example, and should not be construed as limiting the functionality and scope of use of embodiments of the present invention.
As shown in fig. 5, the electronic device 500 is embodied in the form of a general purpose computing device. The components of electronic device 500 may include, but are not limited to: the at least one processing unit 510, the at least one memory unit 520, and a bus 530 connecting the various system components, including the memory unit 520 and the processing unit 510.
Wherein the storage unit stores program code that is executable by the processing unit 510 such that the processing unit 510 performs steps according to various exemplary embodiments of the present invention described in the above section of the "exemplary method" of the present specification. For example, the processing unit 510 may perform step S110 as shown in fig. 1: initializing and generating a plurality of user feature combinations based on the target user feature group; s120: respectively inputting each user feature combination and optimizing target into a first machine learning model trained in advance to obtain an evaluation score of each user feature combination; step S130: acquiring user characteristic combinations corresponding to evaluation scores in a plurality of first score ranges and user characteristic combinations corresponding to evaluation scores in a plurality of second score ranges, wherein the evaluation scores in the plurality of first score ranges are larger than a first preset threshold value, and the evaluation scores in the plurality of second score ranges are lower than a second preset threshold value; step S140: acquiring user characteristic differences of the user characteristic combinations of each first score range and the user characteristic combinations of each second score range, and obtaining a plurality of user characteristic differences; step S150: inputting the multiple user feature distinction and optimization targets into a pre-trained second machine learning model to obtain a predictive optimization user feature combination of the target user feature group; step S160: and acquiring target optimized user feature combinations from the target user feature groups based on the predicted optimized user feature combinations.
The storage unit 520 may include readable media in the form of volatile storage units, such as Random Access Memory (RAM) 5201 and/or cache memory unit 5202, and may further include Read Only Memory (ROM) 5203.
The storage unit 520 may also include a program/utility 5204 having a set (at least one) of program modules 5205, such program modules 5205 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of which may include an implementation of a network environment.
Bus 530 may be one or more of several types of bus structures including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.
The electronic device 500 may also communicate with one or more external devices 700 (e.g., keyboard, pointing device, bluetooth device, etc.), one or more devices that enable a client to interact with the electronic device 500, and/or any device (e.g., router, modem, etc.) that enables the electronic device 500 to communicate with one or more other computing devices. Such communication may occur through an input/output (I/O) interface 550. Also, electronic device 500 may communicate with one or more networks such as a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network, such as the Internet, through network adapter 560. As shown, network adapter 560 communicates with other modules of electronic device 500 over bus 530. It should be appreciated that although not shown, other hardware and/or software modules may be used in connection with electronic device 500, including, but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, data backup storage systems, and the like.
From the above description of embodiments, those skilled in the art will readily appreciate that the example embodiments described herein may be implemented in software, or may be implemented in software in combination with the necessary hardware. Thus, the technical solution according to the embodiments of the present disclosure may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (may be a CD-ROM, a U-disk, a mobile hard disk, etc.) or on a network, including several instructions to cause a computing device (may be a personal computer, a server, a terminal device, or a network device, etc.) to perform the method according to the embodiments of the present disclosure.
In an exemplary embodiment of the present disclosure, a computer-readable storage medium having stored thereon a program product capable of implementing the method described above in the present specification is also provided. In some possible embodiments, the various aspects of the invention may also be implemented in the form of a program product comprising program code for causing a terminal device to carry out the steps according to the various exemplary embodiments of the invention as described in the "exemplary methods" section of this specification, when said program product is run on the terminal device.
Referring to fig. 6, a program product 600 for implementing the above-described method according to an embodiment of the present invention is described, which may employ a portable compact disc read only memory (CD-ROM) and include program code, and may be run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited thereto, and in this document, a readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium would include the following: an electrical connection having one or more wires, a portable disk, a hard disk, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The computer readable signal medium may include a data signal propagated in baseband or as part of a carrier wave with readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A readable signal medium may also be any readable medium that is not a readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Program code for carrying out operations of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the client computing device, partly on the client device, as a stand-alone software package, partly on the client computing device and partly on a remote computing device or entirely on the remote computing device or server. In the case of remote computing devices, the remote computing device may be connected to the client computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., connected via the Internet using an Internet service provider).
Furthermore, the above-described drawings are only schematic illustrations of processes included in the method according to the exemplary embodiment of the present application, and are not intended to be limiting. It will be readily appreciated that the processes shown in the above figures do not indicate or limit the temporal order of these processes. In addition, it is also readily understood that these processes may be performed synchronously or asynchronously, for example, among a plurality of modules.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any adaptations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
Claims (10)
1. A method for optimizing user characteristics in a user characteristic group, comprising:
initializing and generating a plurality of user feature combinations based on the target user feature group;
respectively inputting each user feature combination and optimizing target into a first machine learning model trained in advance to obtain an evaluation score of each user feature combination;
Acquiring user characteristic combinations corresponding to evaluation scores in a plurality of first score ranges and user characteristic combinations corresponding to evaluation scores in a plurality of second score ranges, wherein the evaluation scores in the plurality of first score ranges are larger than a first preset threshold value, and the evaluation scores in the plurality of second score ranges are lower than a second preset threshold value;
acquiring user characteristic differences of the user characteristic combinations of each first score range and the user characteristic combinations of each second score range, and obtaining a plurality of user characteristic differences;
inputting the multiple user feature distinction and optimization targets into a pre-trained second machine learning model to obtain a predictive optimization user feature combination of the target user feature group;
and acquiring target optimized user feature combinations from the target user feature groups based on the predicted optimized user feature combinations.
2. The method of claim 1, wherein initializing the generation of the plurality of user feature combinations based on the target user feature group comprises:
sequentially and respectively acquiring a preset number of user features from the target user feature group, wherein the preset number is more than the preset number in the previous time when each acquisition is performed;
And sequentially and respectively arranging the acquired preset number of user features into user feature strings according to the arrangement rule of the target user feature groups to obtain a plurality of user feature combinations.
3. The method of claim 1, wherein initializing the generation of the plurality of user feature combinations based on the target user feature group comprises:
based on the target user feature group, a predetermined number of user feature combinations are randomly initialized.
4. The method of claim 1, wherein after obtaining a combination of user features corresponding to the evaluation scores in a plurality of first score ranges and a combination of user features corresponding to the evaluation scores in a plurality of second score ranges, wherein the evaluation scores in the plurality of first score ranges are greater than a first predetermined threshold and the evaluation scores in the plurality of second score ranges are less than a second predetermined threshold, the method further comprises:
if the first score range and the second score range do not acquire the corresponding user feature combinations, initializing and generating a preset group of a plurality of user feature combinations based on the target user feature group;
respectively inputting the user feature combinations into a pre-trained first machine learning model to obtain evaluation scores of each user feature combination;
Acquiring user characteristic combinations corresponding to the evaluation scores of a plurality of first score ranges which are larger than a first preset threshold value and user characteristic combinations corresponding to the evaluation scores of a plurality of second score ranges which are lower than a second preset threshold value in all the evaluation scores;
and when all the first score range and the second score range acquire the corresponding user feature combination, ending the acquisition.
5. The method of claim 1, wherein said obtaining a user feature difference for each of said first fractional range of user feature combinations and each of said second fractional range of user feature combinations results in a plurality of user feature differences, comprising:
randomly acquiring a user characteristic combination from the user characteristic combinations in each first score range;
randomly acquiring a user characteristic combination from the user characteristic combinations in each second score range;
and acquiring the user characteristic difference of the randomly acquired one user characteristic combination of each first score range and the randomly acquired one user characteristic combination of each second score range, and obtaining a plurality of user characteristic differences.
6. The method of claim 1, wherein said obtaining a user feature difference for each of said first fractional range of user feature combinations and each of said second fractional range of user feature combinations results in a plurality of user feature differences, comprising:
Acquiring first user characteristic combination elements of the user characteristic combinations of each first score range;
acquiring second user characteristic combination elements of the user characteristic combinations of each second score range;
and obtaining distinguishing features of each first user feature combination element and each second user feature combination element, and obtaining user feature distinction of the user feature combination of each first score range and the user feature combination of each second score range.
7. The method of claim 1, wherein the training method of the first machine learning model comprises:
collecting a user characteristic combination and an optimizing target sample set, wherein each sample in the sample set is calibrated with a corresponding evaluation score in advance;
respectively inputting the input data of each sample in the sample set into a machine learning model to obtain an evaluation score corresponding to each sample;
if the input data with the sample is input into the machine learning model, obtaining that the evaluation score corresponding to the sample is inconsistent with the evaluation score calibrated in advance for the sample, and adjusting the coefficient of the learning model until the evaluation score is consistent with the evaluation score calibrated in advance for the sample;
after the input data of all the samples are input into the machine learning model, the obtained evaluation score corresponding to each sample is consistent with the evaluation score calibrated in advance for each sample, and training is finished.
8. A user feature optimizing apparatus in a user feature group, comprising:
the initialization module is used for initializing and generating a plurality of user feature combinations based on the target user feature groups;
the evaluation module is used for inputting each user characteristic combination and the optimizing target into a first machine learning model trained in advance respectively to obtain an evaluation score of each user characteristic combination;
the first acquisition module is used for acquiring user characteristic combinations corresponding to the evaluation scores in a plurality of first score ranges and user characteristic combinations corresponding to the evaluation scores in a plurality of second score ranges, wherein the evaluation scores in the plurality of first score ranges are larger than a first preset threshold value, and the evaluation scores in the plurality of second score ranges are lower than a second preset threshold value;
the second acquisition module is used for acquiring the user characteristic differences of the user characteristic combination of each first score range and the user characteristic combination of each second score range to obtain a plurality of user characteristic differences;
the prediction module is used for inputting the multiple user feature distinction and optimization targets into a pre-trained second machine learning model to obtain a predicted and optimized user feature combination of the target user feature group;
And the optimized feature acquisition module is used for acquiring the target optimized user feature combination from the target user feature group based on the predicted optimized user feature combination.
9. A computer readable storage medium having stored thereon a user feature optimizing program in a user feature group, wherein the user feature optimizing program in the user feature group, when executed by a processor, implements the method of any of claims 1-7.
10. An electronic device, comprising:
a processor; and
the memory is used for storing a user characteristic optimizing program in the user characteristic group of the processor; wherein the processor is configured to perform the method of any of claims 1-7 via execution of a user feature optimizing program in the user feature group.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910533729.8A CN110348581B (en) | 2019-06-19 | 2019-06-19 | User feature optimizing method, device, medium and electronic equipment in user feature group |
| PCT/CN2019/103437 WO2020252925A1 (en) | 2019-06-19 | 2019-08-29 | Method and apparatus for searching user feature group for optimized user feature, electronic device, and computer nonvolatile readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910533729.8A CN110348581B (en) | 2019-06-19 | 2019-06-19 | User feature optimizing method, device, medium and electronic equipment in user feature group |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110348581A CN110348581A (en) | 2019-10-18 |
| CN110348581B true CN110348581B (en) | 2023-08-18 |
Family
ID=68182439
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910533729.8A Active CN110348581B (en) | 2019-06-19 | 2019-06-19 | User feature optimizing method, device, medium and electronic equipment in user feature group |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN110348581B (en) |
| WO (1) | WO2020252925A1 (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111737319B (en) * | 2020-06-24 | 2024-03-01 | 平安科技（深圳）有限公司 | User cluster prediction method, device, computer equipment and storage medium |
| CN112669982B (en) * | 2020-12-31 | 2023-02-21 | 南方科技大学 | Method, device, equipment and storage medium for determining close contact person |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105260414A (en) * | 2015-09-24 | 2016-01-20 | 精硕世纪科技（北京）有限公司 | User behavior similarity computing method and device |
| CN106548210A (en) * | 2016-10-31 | 2017-03-29 | 腾讯科技（深圳）有限公司 | Machine learning model training method and device |
| CN108154430A (en) * | 2017-12-28 | 2018-06-12 | 上海氪信信息技术有限公司 | A kind of credit scoring construction method based on machine learning and big data technology |
| CN109858974A (en) * | 2019-02-18 | 2019-06-07 | 重庆邮电大学 | Automobile-used family identification model construction method and recognition methods are purchased |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8266078B2 (en) * | 2009-02-06 | 2012-09-11 | Microsoft Corporation | Platform for learning based recognition research |
| CN104239351B (en) * | 2013-06-20 | 2017-12-19 | 阿里巴巴集团控股有限公司 | A kind of training method and device of the machine learning model of user behavior |
| US10509791B2 (en) * | 2016-03-08 | 2019-12-17 | Facebook, Inc. | Statistical feature engineering of user attributes |
| CN107480713A (en) * | 2017-08-08 | 2017-12-15 | 云南大学 | Co location mode excavation methods containing dominant characteristics |
-
2019
- 2019-06-19 CN CN201910533729.8A patent/CN110348581B/en active Active
- 2019-08-29 WO PCT/CN2019/103437 patent/WO2020252925A1/en active Application Filing
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105260414A (en) * | 2015-09-24 | 2016-01-20 | 精硕世纪科技（北京）有限公司 | User behavior similarity computing method and device |
| CN106548210A (en) * | 2016-10-31 | 2017-03-29 | 腾讯科技（深圳）有限公司 | Machine learning model training method and device |
| CN108154430A (en) * | 2017-12-28 | 2018-06-12 | 上海氪信信息技术有限公司 | A kind of credit scoring construction method based on machine learning and big data technology |
| CN109858974A (en) * | 2019-02-18 | 2019-06-07 | 重庆邮电大学 | Automobile-used family identification model construction method and recognition methods are purchased |
Non-Patent Citations (1)
| Title |
|---|
| 特征提取与多目标机器学习研究及应用;何涛等;四川理工学院学报(自然科学版);第26卷(第01期);第39-43页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2020252925A1 (en) | 2020-12-24 |
| CN110348581A (en) | 2019-10-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110995459B (en) | Abnormal object identification method, device, medium and electronic equipment | |
| CN108520470B (en) | Method and apparatus for generating user attribute information | |
| CN109242044A (en) | Training method, device, storage medium and the electronic equipment of vehicle and goods matching model | |
| CN112994701B (en) | Data compression method, device, electronic equipment and computer readable medium | |
| CN111209351B (en) | Object relation prediction method, object recommendation method, object relation prediction device, object recommendation device, electronic equipment and medium | |
| CN110708285B (en) | Flow monitoring method, device, medium and electronic equipment | |
| CN111209478A (en) | Task pushing method and device, storage medium and electronic equipment | |
| CN105069036A (en) | Information recommendation method and apparatus | |
| CN110135769B (en) | Goods attribute filling method and device, storage medium and electronic terminal | |
| CN112883990A (en) | Data classification method and device, computer storage medium and electronic equipment | |
| CN110348581B (en) | User feature optimizing method, device, medium and electronic equipment in user feature group | |
| CN111582645B (en) | APP risk assessment method and device based on factoring machine and electronic equipment | |
| CN117573507A (en) | Test case generation method and device, electronic equipment and storage medium | |
| CN116684330A (en) | Traffic prediction method, device, equipment and storage medium based on artificial intelligence | |
| CN113591881B (en) | Intention recognition method and device based on model fusion, electronic equipment and medium | |
| CN113392920B (en) | Method, apparatus, device, medium, and program product for generating cheating prediction model | |
| CN109272165B (en) | Registration probability estimation method and device, storage medium and electronic equipment | |
| CN113157853B (en) | Problem mining method, device, electronic equipment and storage medium | |
| CN111161238A (en) | Image quality evaluation method and device, electronic device, and storage medium | |
| CN116030115B (en) | Visual design image analysis method and system applied to AI | |
| JP2019101681A (en) | Quotation acquisition device, quotation acquisition method, and program | |
| CN115186738B (en) | Model training method, device and storage medium | |
| CN110879821A (en) | Method, device, equipment and storage medium for generating rating card model derivative label | |
| CN114237182B (en) | Robot scheduling method and system | |
| CN115269998A (en) | Information recommendation method and device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |