CN111738778B - User portrait generation method and device, computer equipment and storage medium - Google Patents
User portrait generation method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN111738778B CN111738778B CN202010695723.3A CN202010695723A CN111738778B CN 111738778 B CN111738778 B CN 111738778B CN 202010695723 A CN202010695723 A CN 202010695723A CN 111738778 B CN111738778 B CN 111738778B
- Authority
- CN
- China
- Prior art keywords
- user
- word segmentation
- time unit
- feature
- result
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Abstract
The invention relates to the technical field of big data, and provides a user portrait generation method, a device, computer equipment and a storage medium, wherein the method comprises the following steps: performing word segmentation on text data of a user in each time unit to obtain word segmentation results and performing discretization on the structured data to obtain discretization results; calculating the personnel type of the user according to the word segmentation result; performing feature screening on the discretization result in the corresponding time unit according to the word segmentation result to obtain a target feature, and calculating the overall score of the target feature by using a gradient lifting tree algorithm; the time factors and the overall scores in each time unit are added and averaged to obtain user ratings; and generating a user portrait according to the personnel type and the user rating. The method can be used in intelligent government affairs, can generate the user portrait by combining the text data and the structured data of the user, and is higher in accuracy. In addition, the invention relates to the technical field of blockchains, and the text data and the structured data can be acquired from the blockchains.
Description
Technical Field
The invention relates to the technical field of big data analysis, in particular to a user portrait generation method and device, computer equipment and a storage medium.
Background
Personnel evaluation is an important basis in personnel management, and plays a very important role in all aspects of personnel record, selection, excavation potential and the like. Under the background of the era of big data, various information of users is filled in a network, each concrete information of the users is abstracted into labels, and the labels are utilized to embody the user image to obtain concrete user portrait, so that the targeted post matching is provided for enterprises based on the user portrait.
At present, a model is trained mainly by combining structured data with a deep learning mode, and the trained model is used for quantitative scoring. However, this scoring method has the following disadvantages: the data source for training the deep learning model is limited, so that the accuracy of the scoring result given by the trained model is not high; secondly, the scoring can only show the final result, and cannot visually and comprehensively depict various abilities of personnel, and the evaluation result is relatively one-sided.
Disclosure of Invention
In view of the foregoing, there is a need for a user representation generation method, apparatus, computer device and storage medium that can combine user text data and structured data to generate a user representation with higher accuracy.
A first aspect of the present invention is a user representation generation method, the method comprising:
acquiring text data and structured data of a user in each time unit;
performing word segmentation processing on the text data to obtain word segmentation results and performing discretization processing on the structured data to obtain discretization results;
calculating the personnel type of the user according to the word segmentation result in each time unit;
performing feature screening on the discretization result in the corresponding time unit according to the word segmentation result to obtain a target feature, and calculating the overall score of the target feature by using a gradient lifting tree algorithm;
the time factors and the overall scores in each time unit are added and averaged to obtain user ratings;
and generating a user portrait according to the personnel type of the user and the user rating.
According to an alternative embodiment of the present invention, the generating a user representation from the user's person types and the user ratings comprises:
carrying out normalization processing on the user rating by adopting a sigmoid function to obtain the final rating of the user;
generating a user portrait according to the personnel type of the user and the final rating;
the sigmoid function is as follows: α + (1/(1+5 × e ^ ((((β - α)/10)))), where α is a benchmark score, as a coefficient, and β is the user rating.
According to an optional embodiment of the present invention, the performing a word segmentation process on the text data to obtain a word segmentation result and performing a discretization process on the structured data to obtain a discretization result includes:
calling a Chinese word segmentation algorithm to segment words of the text data, and removing useless words after word segmentation to obtain word segmentation results;
and comparing each item of structured data in the structured data with a plurality of preset data ranges corresponding to the structured data, and dispersing the structured data in each preset data range to a score corresponding to the preset data range to obtain a discretization result.
According to an optional embodiment of the present invention, the calculating the person type of the user according to the word segmentation result in each time unit comprises:
matching the word segmentation result with a plurality of feature words corresponding to a plurality of preset personnel types in a regularization mode; determining feature words which are matched from a plurality of feature words corresponding to the preset personnel types and have the same word segmentation result as target feature words; calculating the number of target feature words corresponding to each preset personnel type; determining the preset personnel type with the largest number of the target characteristic words as the personnel type of the user; or
Obtaining historical word segmentation results of a plurality of historical users and marking personnel types according to the historical word segmentation results; training a neural network based on the labeled personnel types and the historical word segmentation results to obtain a personnel type identification model; and calling the personnel type identification model to identify the personnel type of the user according to the word segmentation result of the user.
According to an optional embodiment of the present invention, the calculating the person type of the user according to the word segmentation result in each time unit comprises:
generating a word segmentation vector for each word segmentation in the word segmentation result by using a word2vec model;
performing semantic analysis on each participle by using a BilSTM-CRF model and generating a participle weight according to a semantic analysis result;
calculating the similarity between the word segmentation vector of each word segmentation and the word segmentation vector corresponding to the preset personnel type aiming at each preset personnel type, and weighting and calculating the similarity of all the word segmentation and the word segmentation weight of the corresponding word segmentation to obtain the score of the preset personnel type;
and determining the preset personnel type corresponding to the highest score as the personnel type of the user.
According to an optional embodiment of the present invention, the performing feature screening on the discretization result in the corresponding time unit according to the word segmentation result to obtain a target feature, and calculating an overall score of the target feature by using a gradient lifting tree algorithm includes:
taking each word segmentation in the word segmentation result as a feature, and marking a score for the feature;
performing feature screening on the discretization result by adopting a random forest algorithm to obtain target features;
and calculating an overall score based on the target characteristics and the corresponding labeling scores by adopting a gradient lifting tree algorithm.
According to an alternative embodiment of the invention, the method further comprises:
acquiring post information;
matching a user portrait corresponding to the post information;
judging the number of matched user images;
when the number is 1, recommending the user corresponding to the matched user image to a manager of the post information;
and when the number is larger than 1, acquiring the final rating in each user portrait, and recommending the user corresponding to the user portrait with the highest final rating to the manager of the position information.
A second aspect of the present invention is a user representation generating apparatus, comprising:
the data acquisition module is used for acquiring text data and structured data of a user in each time unit;
the data processing module is used for performing word segmentation processing on the text data to obtain word segmentation results and performing discretization processing on the structured data to obtain discretization results;
the type calculation module is used for calculating the personnel type of the user according to the word segmentation result in each time unit;
the score calculation module is used for carrying out feature screening on the discretization result in the corresponding time unit according to the word segmentation result to obtain a target feature, and calculating the overall score of the target feature by using a gradient lifting tree algorithm;
the user rating module is used for carrying out addition and average calculation on the time factor and the integral score in each time unit to obtain user rating;
and the portrait generation module is used for generating a user portrait according to the personnel type of the user and the user rating.
A third aspect of the invention provides a computer apparatus comprising a processor for implementing the user representation generation method when executing a computer program stored in a memory.
A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the user representation generation method.
In summary, the user portrait generation method, the user portrait generation device, the computer device and the storage medium of the present invention obtain the information of the user in the two aspects of text data and structured data more comprehensively and accurately, and the user portrait generated based on the text data and the structured data has higher accuracy, thereby improving the accuracy of the user portrait. The user portrait generation method is applied to post recruitment, and the most suitable person selection can be accurately matched for the post.
Drawings
FIG. 1 is a flowchart of a user representation generation method according to an embodiment of the present invention.
FIG. 2 is a block diagram of a user representation generating apparatus according to a second embodiment of the present invention.
Fig. 3 is a schematic structural diagram of a computer device according to a third embodiment of the present invention.
Detailed Description
In order that the above objects, features and advantages of the present invention can be more clearly understood, a detailed description of the present invention will be given below with reference to the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present invention and features of the embodiments may be combined with each other without conflict.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.
FIG. 1 is a flowchart of a user representation generation method according to an embodiment of the present invention. The user representation generation method is executed by computer equipment and specifically comprises the following steps, and the sequence of the steps in the flow chart can be changed and some steps can be omitted according to different requirements.
And S11, acquiring the text data and the structured data of the user in each time unit.
The computer device may preset time units and obtain text data and structured data related to the user in each time unit. Generally, enterprises set up a plurality of assessment targets to conduct annual or quarterly or monthly assessment on employees, and assessment results corresponding to the assessment targets are stored in a database in the form of text data and/or structured data. The computer device searches the text data and the structured data of the user from the database with the time unit as a search condition.
The time unit may be a year, half a year, quarter, month, or the like.
The text data is text type data, including but not limited to: self-evaluation text data, others evaluation text data.
The structured data is voting data, including but not limited to: the latitude information comprises evaluation of other people, negative information, task completion condition, work experience, academic calendar, titles, project practice information and the like. Each latitude information corresponds to one structured data. For example, the evaluation of the latitude by others corresponds to a figure with excellent evaluation, a figure with good evaluation and a figure with poor evaluation.
And S12, performing word segmentation processing on the text data to obtain word segmentation results and performing discretization processing on the structured data to obtain discretization results.
After the text data and the structured data are obtained by the computer equipment, the text data and the structured data are processed firstly and serve as a data source for subsequently calculating the user capability rating.
In an optional embodiment, the performing a word segmentation process on the text data to obtain a word segmentation result and performing a discretization process on the structured data to obtain a discretization result includes:
calling a Chinese word segmentation algorithm to segment words of the text data, and removing useless words after word segmentation to obtain word segmentation results;
and comparing each item of structured data in the structured data with a plurality of preset data ranges corresponding to the structured data, and dispersing the structured data in each preset data range to a score corresponding to the preset data range to obtain a discretization result.
In this optional embodiment, the chinese word segmentation algorithm may be a maximum matching word segmentation algorithm, a hidden markov algorithm, or the like, performs chinese word segmentation on each text data in each time unit, and removes stop words, numbers, symbols, or other useless words after word segmentation to obtain a word segmentation result, so as to facilitate subsequent calculation of the personnel type based on the word segmentation result.
Discretization processing is carried out on each item of structured data in the structured data to obtain a discretization result, so that a plurality of structured data are conveniently mapped to one reference value, the data volume of the structured data is reduced, and the efficiency of subsequent feature screening is conveniently improved. For example, assuming that the work experience of the user is 4 years, the plurality of preset data ranges corresponding to the work experience are: [0, 1], [1, 3], [3, 5], [5, 10], [10, 10+ ], and the score corresponding to each preset data range is respectively: 1, 2, 3, 4, 5, the work experience of the user is discretized into a score of 3.
And S13, calculating the personnel type of the user according to the word segmentation result in each time unit.
The person type can represent to some extent the personality of the user and the success of the work. As employment experiences grow and the personnel types of the user can change greatly, the computer equipment can calculate the personnel types of the user in each time unit according to a plurality of word segmentation results, so that the change track of the personnel types of the user can be conveniently determined, and therefore, the proper position is arranged for the personnel according to the personnel types.
In an optional embodiment, the calculating the person type of the user according to the word segmentation result in each time unit includes:
matching the word segmentation result with a plurality of feature words corresponding to a plurality of preset personnel types in a regularization mode;
determining feature words which are matched from a plurality of feature words corresponding to the preset personnel types and have the same word segmentation results as target feature words;
calculating the number of target feature words corresponding to each preset personnel type;
and determining the preset personnel type with the maximum number of the target characteristic words as the personnel type of the user.
In this alternative embodiment, a plurality of person types, for example, a tiger type, a peacock type, a caraway type, and a owl type, are stored in advance in the computer device, and a plurality of feature words are provided for each person type in order to describe the features of each person type quantitatively. The person type and corresponding feature words are denoted C1, C2, …, Cn, e.g., peacock type person type { chinese don't see, middle-sighted, …, good at the junction }.
Matching each word segmentation result with each feature word corresponding to each preset personnel type in a regularization mode, considering that the word segmentation is hit when any feature word in any personnel type is successfully matched with a word segmentation, and counting and accumulating for 1; and when all the characteristic words of all the personnel types are failed to be matched with a certain participle, the participle is considered to be not hit.
In an alternative embodiment, the calculating the person type of the user according to the word segmentation result in each time unit includes:
generating a word segmentation vector for each word segmentation in the word segmentation result by using a word2vec model;
performing semantic analysis on each participle by using a BilSTM-CRF model and generating a participle weight according to a semantic analysis result; calculating the similarity between the word segmentation vector of each word segmentation and the word segmentation vector corresponding to the preset personnel type aiming at each preset personnel type, and weighting and calculating the similarity of all the word segmentation and the word segmentation weight of the corresponding word segmentation to obtain the score of the preset personnel type;
and determining the preset personnel type corresponding to the highest score as the personnel type of the user.
In this alternative embodiment, Word2vec is a cluster of correlation models used to generate Word vectors, and the Word2vec model may be used to map each Word to a vector. The BilSTM-CRF model is a model based on a BI-directional long-short-term memory (BI-LSTM) neural network and a Conditional Random Field (CRF), and is used for analyzing the part of speech of a participle to determine whether a person corresponding to the participle is an actor or a recipient, so as to generate different weights. Exemplarily, assuming that semantic parsing actively and efficiently completes a project for a certain user, it indicates that the user is a donor and generates a larger weight; assuming that semantic parsing is such that a user efficiently follows a command to complete a project, it indicates that the user is the recipient, and generates a smaller weight. The sum of the generated weights is 1.
The word2vec model and the BilSTM-CRF model are prior art and the invention will not be elaborated upon here.
In another alternative embodiment, said calculating the person type of the user according to the word segmentation result in each time unit comprises:
obtaining historical word segmentation results of a plurality of historical users and marking personnel types according to the historical word segmentation results;
training a neural network based on the labeled personnel types and the historical word segmentation results to obtain a personnel type identification model;
and calling the personnel type identification model to identify the personnel type of the user according to the word segmentation result of the user.
In this alternative embodiment, a person type recognition model is trained based on the person types and word segmentation results of the users who have been in history as training data to predict the person types of the users. The process of training the person type recognition model is prior art and the present invention is not described in detail.
S14, performing feature screening on the discretization result in the corresponding time unit according to the word segmentation result to obtain a target feature, and calculating the overall score of the target feature by using a gradient lifting tree algorithm.
The discretization result comprises data of a plurality of latitudes, and the data of the plurality of latitudes do not influence the calculation of the rating, so that data which has a larger influence on the calculation result of the rating needs to be selected from the data.
In an optional embodiment, the performing feature screening on the discretization result in the corresponding time unit according to the word segmentation result to obtain a target feature, and calculating an overall score of the target feature by using a gradient lifting tree algorithm includes:
taking each word segmentation in the word segmentation result as a feature, and marking a score for the feature;
performing feature screening on the discretization result by adopting a random forest algorithm to obtain target features;
and calculating an overall score based on the target characteristics and the corresponding labeling scores by adopting a gradient lifting tree algorithm.
In this alternative embodiment, the random forest algorithm mainly looks at how much each feature contributes to each tree in the random forest, then averages the contributions, and finally compares the contributions between different features. The greater the contribution, the more important the feature is, the need to be preserved; the smaller the contribution, the less important the feature is, the need for culling. And constructing a random forest tree for the discretization result by adopting a random forest algorithm, and determining the characteristics corresponding to the trees larger than a preset characteristic threshold value as target characteristics.
The Gradient Boosting Decision Tree (GBDT) algorithm is a Boosting method that uses an additive model (i.e., a linear combination of basis functions) and a forward step-by-step algorithm, and uses a Decision Tree as a basis function. The algorithm consists of a plurality of decision trees, and the conclusions of all the trees are added to form a final answer.
And S15, adding and averaging the time factors in each time unit and the overall score to obtain the user rating.
The computer device has pre-stored therein a time factor, for example, 0.318 x (-0.125), where x is year and grows geometrically in multiples, the first year is taken to be 0.0005, the second year is 2 and starts to grow in multiples of 2, i.e., the third year is 4, the fourth year is 8, and so on.
And sequentially calculating the time factor and the overall score of the first year to obtain a first user rating, calculating the time factor and the overall score of the second year to obtain a second user rating, and repeating the steps until all the user ratings are summed and then averaged to obtain the user rating of the user.
And S16, generating a user portrait according to the personnel type of the user and the user rating.
A user representation generated by a computer device includes the user's person type and user ratings, structured data, and the like.
In an alternative embodiment, said generating a user representation from said user's person type and said user rating comprises:
carrying out normalization processing on the user rating by adopting a sigmoid function to obtain the final rating of the user;
and generating a user portrait according to the personnel type of the user and the final rating.
In this optional embodiment, the interval range of the person rating is set in advance in the computer device, and therefore, after the user rating of the user is calculated according to the time factor, the user rating needs to be normalized by using a sigmoid function, so as to obtain a final rating.
In an alternative embodiment, the sigmoid function may be: α + (1/(1+5 × e ^ ((((β - α)/10)))), where α is a benchmark score, as a coefficient, and β is the user rating.
The final rating is obtained by normalizing the user rating, so that the ratings of all users can be normalized to a uniform interval conveniently, the uniform price standard is achieved, and the obtained final rating of the user has a more referential meaning.
In an optional embodiment, the method further comprises:
acquiring post information;
matching a user portrait corresponding to the post information;
judging the number of matched user images;
when the number is 1, recommending the user corresponding to the matched user image to a manager of the post information;
and when the number is larger than 1, acquiring the final rating in each user portrait, and recommending the user corresponding to the user portrait with the highest final rating to the manager of the position information.
In the optional embodiment, if the enterprise publishes the post information and recruits a proper employee from the interior of the enterprise, the type of the employee corresponding to the service type can be matched according to the service type of the post information, and if the number of the matched user figures is large, the preferential selection is performed according to the final rating.
For example, assuming that the service type in the published position information is market development, a user portrait with a peacock type person type may be matched, and if three user portraits are matched, the final ratings of the three user portraits are: 90, 85, 89. And selecting the user corresponding to the user image with the final rating of 90 as the recommending personnel of the position information.
In conclusion, the information obtained by acquiring the text data and the structured data of the user is more comprehensive and accurate, the user portrait generated based on the text data and the structured data is higher in accuracy, and the accuracy of the user portrait is improved. The user portrait generation method of the embodiment can be applied to intelligent government affairs to promote the construction of intelligent cities. For example, the user representation generation method according to the embodiment is applied to post recruitment, and can accurately match posts to obtain the most suitable person selection.
It should be noted that, in order to ensure the privacy and security of the text data and the structured data of the user, the text data and the structured data of the user may be obtained from the block link point.
FIG. 2 is a block diagram of a user representation generating apparatus according to a second embodiment of the present invention.
In some embodiments, the user representation generating device 20 may comprise a plurality of functional modules comprising computer program segments. Computer programs for the various program segments of the user representation generation apparatus 20 may be stored in a memory of a computer device and executed by at least one processor to perform the functions of user representation generation (described in detail in FIG. 1).
In this embodiment, the user representation generating apparatus 20 may be divided into a plurality of functional blocks according to the functions to be executed. The functional module may include: a data acquisition module 201, a data processing module 202, a type calculation module 203, a score calculation module 204, a user rating module 205, a representation generation module 206, a post acquisition module 207, and a post recommendation module 208. The module referred to herein is a series of computer program segments capable of being executed by at least one processor and capable of performing a fixed function and is stored in memory. In the present embodiment, the functions of the modules will be described in detail in the following embodiments.
The data obtaining module 201 is configured to obtain text data and structured data of a user in each time unit.
The computer device may preset time units and obtain text data and structured data related to the user in each time unit. Generally, enterprises set up a plurality of assessment targets to conduct annual or quarterly or monthly assessment on employees, and assessment results corresponding to the assessment targets are stored in a database in the form of text data and/or structured data. The computer device searches the text data and the structured data of the user from the database with the time unit as a search condition.
The time unit may be a year, half a year, quarter, month, or the like.
The text data is text type data, including but not limited to: self-evaluation text data, others evaluation text data.
The structured data is voting data, including but not limited to: the latitude information comprises evaluation of other people, negative information, task completion condition, work experience, academic calendar, titles, project practice information and the like. Each latitude information corresponds to one structured data. For example, the evaluation of the latitude by others corresponds to a figure with excellent evaluation, a figure with good evaluation and a figure with poor evaluation.
The data processing module 202 is configured to perform word segmentation on the text data to obtain a word segmentation result, and perform discretization on the structured data to obtain a discretization result.
After the text data and the structured data are obtained by the computer equipment, the text data and the structured data are processed firstly and serve as a data source for subsequently calculating the user capability rating.
In an optional embodiment, the data processing module 202 performs word segmentation on the text data to obtain a word segmentation result, and performs discretization on the structured data to obtain a discretization result, including:
calling a Chinese word segmentation algorithm to segment words of the text data, and removing useless words after word segmentation to obtain word segmentation results;
and comparing each item of structured data in the structured data with a plurality of preset data ranges corresponding to the structured data, and dispersing the structured data in each preset data range to a score corresponding to the preset data range to obtain a discretization result.
In this optional embodiment, the chinese word segmentation algorithm may be a maximum matching word segmentation algorithm, a hidden markov algorithm, or the like, performs chinese word segmentation on each text data in each time unit, and removes stop words, numbers, symbols, or other useless words after word segmentation to obtain a word segmentation result, so as to facilitate subsequent calculation of the personnel type based on the word segmentation result.
Discretization processing is carried out on each item of structured data in the structured data to obtain a discretization result, so that a plurality of structured data are conveniently mapped to one reference value, the data volume of the structured data is reduced, and the efficiency of subsequent feature screening is conveniently improved. For example, assuming that the work experience of the user is 4 years, the plurality of preset data ranges corresponding to the work experience are: [0, 1], [1, 3], [3, 5], [5, 10], [10, 10+ ], and the score corresponding to each preset data range is respectively: 1, 2, 3, 4, 5, the work experience of the user is discretized into a score of 3.
The type calculating module 203 is configured to calculate a person type of the user according to the word segmentation result in each time unit.
The person type can represent to some extent the personality of the user and the success of the work. As employment experiences grow and the personnel types of the user can change greatly, the computer equipment can calculate the personnel types of the user in each time unit according to a plurality of word segmentation results, so that the change track of the personnel types of the user can be conveniently determined, and therefore, the proper position is arranged for the personnel according to the personnel types.
In an alternative embodiment, the calculating the person type of the user according to the word segmentation result in each time unit by the type calculating module 203 includes:
matching the word segmentation result with a plurality of feature words corresponding to a plurality of preset personnel types in a regularization mode;
determining feature words which are matched from a plurality of feature words corresponding to the preset personnel types and have the same word segmentation results as target feature words;
calculating the number of target feature words corresponding to each preset personnel type;
and determining the preset personnel type with the maximum number of the target characteristic words as the personnel type of the user.
In this alternative embodiment, a plurality of person types, for example, a tiger type, a peacock type, a caraway type, and a owl type, are stored in advance in the computer device, and a plurality of feature words are provided for each person type in order to describe the features of each person type quantitatively. The person type and corresponding feature words are denoted C1, C2, …, Cn, e.g., peacock type person type { chinese don't see, middle-sighted, …, good at the junction }.
Matching each word segmentation result with each feature word corresponding to each preset personnel type in a regularization mode, considering that the word segmentation is hit when any feature word in any personnel type is successfully matched with a word segmentation, and counting and accumulating for 1; and when all the characteristic words of all the personnel types are failed to be matched with a certain participle, the participle is considered to be not hit.
In an alternative embodiment, the calculating the person type of the user by the type calculating module 203 according to the word segmentation result in each time unit comprises:
generating a word segmentation vector for each word segmentation in the word segmentation result by using a word2vec model;
performing semantic analysis on each participle by using a BilSTM-CRF model and generating a participle weight according to a semantic analysis result;
calculating the similarity between the word segmentation vector of each word segmentation and the word segmentation vector corresponding to the preset personnel type aiming at each preset personnel type, and weighting and calculating the similarity of all the word segmentation and the word segmentation weight of the corresponding word segmentation to obtain the score of the preset personnel type;
and determining the preset personnel type corresponding to the highest score as the personnel type of the user.
In this alternative embodiment, Word2vec is a cluster of correlation models used to generate Word vectors, and the Word2vec model may be used to map each Word to a vector. The BilSTM-CRF model is a model based on a BI-directional long-short-term memory (BI-LSTM) neural network and a Conditional Random Field (CRF), and is used for analyzing the part of speech of a participle to determine whether a person corresponding to the participle is an actor or a recipient, so as to generate different weights. Exemplarily, assuming that semantic parsing actively and efficiently completes a project for a certain user, it indicates that the user is a donor and generates a larger weight; assuming that semantic parsing is such that a user efficiently follows a command to complete a project, it indicates that the user is the recipient, and generates a smaller weight. The sum of the generated weights is 1.
The word2vec model and the BilSTM-CRF model are prior art and the invention will not be elaborated upon here.
In another alternative embodiment, the calculating the person type of the user by the type calculating module 203 according to the word segmentation result in each time unit comprises:
obtaining historical word segmentation results of a plurality of historical users and marking personnel types according to the historical word segmentation results;
training a neural network based on the labeled personnel types and the historical word segmentation results to obtain a personnel type identification model;
and calling the personnel type identification model to identify the personnel type of the user according to the word segmentation result of the user.
In this alternative embodiment, a person type recognition model is trained based on the person types and word segmentation results of the users who have been in history as training data to predict the person types of the users. The process of training the person type recognition model is prior art and the present invention is not described in detail.
The score calculation module 204 is configured to perform feature screening on the discretization result in the corresponding time unit according to the word segmentation result to obtain a target feature, and calculate an overall score of the target feature by using a gradient lifting tree algorithm.
The discretization result comprises data of a plurality of latitudes, and the data of the plurality of latitudes do not influence the calculation of the rating, so that data which has a larger influence on the calculation result of the rating needs to be selected from the data.
In an optional embodiment, the score calculating module 204 performs feature screening on the discretization result in the corresponding time unit according to the word segmentation result to obtain a target feature, and calculating the overall score of the target feature by using a gradient lifting tree algorithm includes:
taking each word segmentation in the word segmentation result as a feature, and marking a score for the feature;
performing feature screening on the discretization result by adopting a random forest algorithm to obtain target features;
and calculating an overall score based on the target characteristics and the corresponding labeling scores by adopting a gradient lifting tree algorithm.
In this alternative embodiment, the random forest algorithm mainly looks at how much each feature contributes to each tree in the random forest, then averages the contributions, and finally compares the contributions between different features. The greater the contribution, the more important the feature is, the need to be preserved; the smaller the contribution, the less important the feature is, the need for culling. And constructing a random forest tree for the discretization result by adopting a random forest algorithm, and determining the characteristics corresponding to the trees larger than a preset characteristic threshold value as target characteristics.
The Gradient Boosting Decision Tree (GBDT) algorithm is a Boosting method that uses an additive model (i.e., a linear combination of basis functions) and a forward step-by-step algorithm, and uses a Decision Tree as a basis function. The algorithm consists of a plurality of decision trees, and the conclusions of all the trees are added to form a final answer.
The user rating module 205 is configured to add and average the time factor and the overall score in each time unit to obtain a user rating.
The computer device has pre-stored therein a time factor, for example, 0.318 x (-0.125), where x is year and grows geometrically in multiples, the first year is taken to be 0.0005, the second year is 2 and starts to grow in multiples of 2, i.e., the third year is 4, the fourth year is 8, and so on.
And sequentially calculating the time factor and the overall score of the first year to obtain a first user rating, calculating the time factor and the overall score of the second year to obtain a second user rating, and repeating the steps until all the user ratings are summed and then averaged to obtain the user rating of the user.
The representation generation module 206 is configured to generate a user representation according to the user's person type and the user rating.
A user representation generated by a computer device includes the user's person type and user ratings, structured data, and the like.
In an alternative embodiment, the representation generation module 206 generating a user representation based on the user's person type and the user rating includes:
carrying out normalization processing on the user rating by adopting a sigmoid function to obtain the final rating of the user;
and generating a user portrait according to the personnel type of the user and the final rating.
In this optional embodiment, the interval range of the person rating is set in advance in the computer device, and therefore, after the user rating of the user is calculated according to the time factor, the user rating needs to be normalized by using a sigmoid function, so as to obtain a final rating.
In an alternative embodiment, the sigmoid function may be: α + (1/(1+5 × e ^ ((((β - α)/10)))), where α is a benchmark score, as a coefficient, and β is the user rating.
The final rating is obtained by normalizing the user rating, so that the ratings of all users can be normalized to a uniform interval conveniently, the uniform price standard is achieved, and the obtained final rating of the user has a more referential meaning.
The post obtaining module 207 is configured to obtain post information and match a user portrait corresponding to the post information.
The post recommending module 208 is configured to determine the number of the matched user images; when the number is 1, recommending the user corresponding to the matched user image to a manager of the post information; and when the number is larger than 1, acquiring the final rating in each user portrait, and recommending the user corresponding to the user portrait with the highest final rating to the manager of the position information.
In the optional embodiment, if the enterprise publishes the post information and recruits a proper employee from the interior of the enterprise, the type of the employee corresponding to the service type can be matched according to the service type of the post information, and if the number of the matched user figures is large, the preferential selection is performed according to the final rating.
For example, assuming that the service type in the published position information is market development, a user portrait with a peacock type person type may be matched, and if three user portraits are matched, the final ratings of the three user portraits are: 90, 85, 89. And selecting the user corresponding to the user image with the final rating of 90 as the recommending personnel of the position information.
In conclusion, the information obtained by acquiring the text data and the structured data of the user is more comprehensive and accurate, the user portrait generated based on the text data and the structured data is higher in accuracy, and the accuracy of the user portrait is improved.
The user portrait generation device of this embodiment can be applied to the intelligent government affairs, promotes the construction of wisdom city. For example, the user representation generation method according to the embodiment is applied to post recruitment, and can accurately match posts to obtain the most suitable person selection.
It should be noted that, in order to ensure the privacy and security of the text data and the structured data of the user, the text data and the structured data of the user may be obtained from the block link point.
Fig. 3 is a schematic structural diagram of a computer device according to a third embodiment of the present invention. In the preferred embodiment of the present invention, the computer device 3 includes a memory 31, at least one processor 32, at least one communication bus 33, and a transceiver 34.
It will be appreciated by those skilled in the art that the configuration of the computer device shown in fig. 3 does not constitute a limitation of the embodiments of the present invention, and may be a bus-type configuration or a star-type configuration, and that the computer device 3 may include more or less hardware or software than those shown, or a different arrangement of components.
In some embodiments, the computer device 3 is a computer device capable of automatically performing numerical calculation and/or information processing according to instructions set or stored in advance, and the hardware thereof includes but is not limited to a microprocessor, an application specific integrated circuit, a programmable gate array, a digital processor, an embedded device, and the like. The computer device 3 may also include a client device, which includes, but is not limited to, any electronic product capable of interacting with a client through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device, for example, a personal computer, a tablet computer, a smart phone, a digital camera, etc.
It should be noted that the computer device 3 is only an example, and other electronic products that are currently available or may come into existence in the future, such as electronic products that can be adapted to the present invention, should also be included in the scope of the present invention, and are included herein by reference.
In some embodiments, a computer program is stored in the memory 31, and the at least one processor 32 may call the computer program stored in the memory 31 to perform the related functions. For example, the respective modules described in the above embodiments are computer programs stored in the memory 31 and executed by the at least one processor 32, thereby implementing the functions of the respective modules. The Memory 31 includes a Read-Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Programmable Read-Only Memory (EPROM), a One-time Programmable Read-Only Memory (OTPROM), an electronically Erasable rewritable Read-Only Memory (Electrically-Erasable Programmable Read-Only Memory (EEPROM)), an optical Read-Only disk (CD-ROM) or other optical disk Memory, a magnetic disk Memory, a tape Memory, or any other medium readable by a computer capable of carrying or storing data.
In some embodiments, the at least one processor 32 is a Control Unit (Control Unit) of the computer device 3, connects various components of the entire computer device 3 by using various interfaces and lines, and executes various functions and processes data of the computer device 3 by running or executing programs or modules stored in the memory 31 and calling data stored in the memory 31. For example, the at least one processor 32, when executing the computer program stored in the memory, implements all or a portion of the steps of the user representation generation method described in embodiments of the present invention. The at least one processor 32 may be composed of an integrated circuit, for example, a single packaged integrated circuit, or may be composed of a plurality of integrated circuits packaged with the same or different functions, including one or more Central Processing Units (CPUs), microprocessors, digital Processing chips, graphics processors, and combinations of various control chips.
In some embodiments, the at least one communication bus 33 is arranged to enable connection communication between the memory 31 and the at least one processor 32 or the like.
Although not shown, the computer device 3 may further include a power supply (such as a battery) for supplying power to each component, and preferably, the power supply may be logically connected to the at least one processor 32 through a power management device, so as to implement functions of managing charging, discharging, and power consumption through the power management device. The power supply may also include any component of one or more dc or ac power sources, recharging devices, power failure detection circuitry, power converters or inverters, power status indicators, and the like. The computer device 3 may further include various sensors, a bluetooth module, a Wi-Fi module, and the like, which are not described herein again.
Further, the computer-readable 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 according to the use of the blockchain node, and the like.
The block chain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, a consensus mechanism, an encryption algorithm and the like. A block chain (Blockchain), which is essentially a decentralized database, is a series of data blocks associated by using a cryptographic method, and each data block contains information of a batch of network transactions, so as to verify the validity (anti-counterfeiting) of the information and generate a next block. The blockchain may include a blockchain underlying platform, a platform product service layer, an application service layer, and the like.
The integrated unit implemented in the form of a software functional module may be stored in a computer-readable storage medium. The software functional module is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a network device, or the like) or a processor (processor) to execute parts of the method for optimizing the network-wide traffic scheduling according to the embodiments of the present invention.
An embodiment of the present invention further provides a storage medium, where the storage medium stores a computer program, and the computer program, when executed by a processor, implements the steps in the user representation generation method embodiment, for example, S11-S16 shown in fig. 1:
s11, acquiring text data and structured data of the user in each time unit;
s12, performing word segmentation processing on the text data to obtain word segmentation results and performing discretization processing on the structured data to obtain discretization results;
s13, calculating the personnel type of the user according to the word segmentation result in each time unit;
s14, performing feature screening on the discretization result in the corresponding time unit according to the word segmentation result to obtain a target feature, and calculating the overall score of the target feature by using a gradient lifting tree algorithm;
s15, adding and averaging the time factors and the overall scores in each time unit to obtain user ratings;
and S16, generating a user portrait according to the personnel type of the user and the user rating.
Alternatively, the computer program, when executed by the processor, implements the functions of the modules in the above device embodiments, such as the module 201 and 208 in fig. 2:
the data acquisition module 201 is configured to acquire text data and structured data of a user in each time unit;
the data processing module 202 is configured to perform word segmentation on the text data to obtain a word segmentation result, and perform discretization on the structured data to obtain a discretization result;
the type calculating module 203 is configured to calculate a person type of the user according to the word segmentation result in each time unit; the score calculation module 204 is configured to perform feature screening on the discretization result in the corresponding time unit according to the word segmentation result to obtain a target feature, and calculate an overall score of the target feature by using a gradient lifting tree algorithm;
the user rating module 205 is configured to add and average the time factor and the overall score in each time unit to obtain a user rating;
the representation generation module 206, configured to generate a user representation according to the user's person type and the user rating;
the post acquisition module 207 is used for acquiring post information and matching a user portrait corresponding to the post information; the post recommending module 208 is configured to determine the number of the matched user images; when the number is 1, recommending the user corresponding to the matched user image to a manager of the post information; and when the number is larger than 1, acquiring the final rating in each user portrait, and recommending the user corresponding to the user portrait with the highest final rating to the manager of the position information.
The integrated unit implemented in the form of a software functional module may be stored in a computer-readable storage medium. The software functional module is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a computer device, or a network device) or a processor (processor) to execute parts of the methods according to the embodiments of the present invention.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is only one logical functional division, and other divisions may be realized in practice.
The modules described as separate parts may or may not be physically separate, and parts displayed as modules may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.
In addition, functional modules in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional module.
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 attributes 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 sign in a claim should not be construed as limiting the claim concerned. Furthermore, it is obvious that the word "comprising" does not exclude other elements or that the singular does not exclude the plural. A plurality of units or means recited in the apparatus claims may also be implemented by one unit or means in software or hardware. The terms first, second, etc. are used to denote names, but not any particular order.
Finally, it should be noted that the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.
Claims (9)
1. A method of user representation generation, the method comprising:
acquiring text data and structured data of a user in each time unit;
performing word segmentation processing on the text data to obtain word segmentation results and performing discretization processing on the structured data to obtain discretization results;
calculating the personnel type of the user according to the word segmentation result in each time unit;
performing feature screening on the discretization result in the corresponding time unit according to the word segmentation result to obtain a target feature, calculating an overall score of the target feature by using a gradient lifting tree algorithm, performing feature screening on the discretization result in the corresponding time unit according to the word segmentation result to obtain the target feature, and calculating the overall score of the target feature by using the gradient lifting tree algorithm includes: taking each word segmentation in the word segmentation result as a feature, and marking a score for the feature; performing feature screening on the discretization result by adopting a random forest algorithm to obtain target features; calculating an overall score based on the target characteristics and the corresponding labeling scores by adopting a gradient lifting tree algorithm;
calculating to obtain the user rating corresponding to each time unit according to the time factor in each time unit and the overall score, and
summing and averaging the user ratings corresponding to all time units to obtain the user rating of the user;
and generating a user portrait according to the personnel type of the user and the user rating.
2. A user representation generation method in accordance with claim 1, wherein said generating a user representation based on said user's person type and said user rating comprises:
carrying out normalization processing on the user rating by adopting a sigmoid function to obtain the final rating of the user;
generating a user portrait according to the personnel type of the user and the final rating;
the sigmoid function is as follows: α + (1/(1+5 × e ^ ((((β - α)/10)))), where α is a benchmark score, as a coefficient, and β is the user rating.
3. The method of generating a user representation as claimed in claim 1, wherein said segmenting said text data into segmented results and discretizing said structured data into discretized results comprises:
calling a Chinese word segmentation algorithm to segment words of the text data, and removing useless words after word segmentation to obtain word segmentation results;
and comparing each item of structured data in the structured data with a plurality of preset data ranges corresponding to the structured data, and dispersing the structured data in each preset data range to a score corresponding to the preset data range to obtain a discretization result.
4. The user representation generation method of claim 1, wherein said calculating a person type of the user from the segmentation results in each time unit comprises:
matching the word segmentation result with a plurality of feature words corresponding to a plurality of preset personnel types in a regularization mode; determining feature words which are matched from a plurality of feature words corresponding to the preset personnel types and have the same word segmentation result as target feature words; calculating the number of target feature words corresponding to each preset personnel type; determining the preset personnel type with the largest number of the target characteristic words as the personnel type of the user; or
Obtaining historical word segmentation results of a plurality of historical users and marking personnel types according to the historical word segmentation results; training a neural network based on the labeled personnel types and the historical word segmentation results to obtain a personnel type identification model; and calling the personnel type identification model to identify the personnel type of the user according to the word segmentation result of the user.
5. The user representation generation method of claim 1, wherein said calculating a person type of the user from the segmentation results in each time unit comprises:
generating a word segmentation vector for each word segmentation in the word segmentation result by using a word2vec model;
performing semantic analysis on each participle by using a BilSTM-CRF model and generating a participle weight according to a semantic analysis result;
calculating the similarity between the word segmentation vector of each word segmentation and the word segmentation vector corresponding to the preset personnel type aiming at each preset personnel type, and weighting and calculating the similarity of all the word segmentation and the word segmentation weight of the corresponding word segmentation to obtain the score of the preset personnel type;
and determining the preset personnel type corresponding to the highest score as the personnel type of the user.
6. A user representation generation method as claimed in claim 1, the method further comprising:
acquiring post information;
matching a user portrait corresponding to the post information;
judging the number of matched user images;
when the number is 1, recommending the user corresponding to the matched user image to a manager of the post information;
and when the number is larger than 1, acquiring the final rating in each user portrait, and recommending the user corresponding to the user portrait with the highest final rating to the manager of the position information.
7. A user representation generation apparatus, the apparatus comprising:
the data acquisition module is used for acquiring text data and structured data of a user in each time unit;
the data processing module is used for performing word segmentation processing on the text data to obtain word segmentation results and performing discretization processing on the structured data to obtain discretization results;
the type calculation module is used for calculating the personnel type of the user according to the word segmentation result in each time unit;
the score calculation module is configured to perform feature screening on the discretization result in the corresponding time unit according to the word segmentation result to obtain a target feature, calculate an overall score of the target feature by using a gradient lifting tree algorithm, perform feature screening on the discretization result in the corresponding time unit according to the word segmentation result to obtain the target feature, and calculate the overall score of the target feature by using the gradient lifting tree algorithm, where the feature screening includes: taking each word segmentation in the word segmentation result as a feature, and marking a score for the feature; performing feature screening on the discretization result by adopting a random forest algorithm to obtain target features; calculating an overall score based on the target characteristics and the corresponding labeling scores by adopting a gradient lifting tree algorithm;
the user rating module is used for calculating to obtain the user rating corresponding to each time unit according to the time factor in each time unit and the integral score, summing the user ratings corresponding to all the time units and averaging to obtain the user rating of the user;
and the portrait generation module is used for generating a user portrait according to the personnel type of the user and the user rating.
8. A computer device, characterized in that the computer device comprises a processor for implementing a user representation generation method as claimed in any one of claims 1 to 6 when executing a computer program stored in a memory.
9. A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements a user representation generation method as claimed in any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010695723.3A CN111738778B (en) | 2020-07-20 | 2020-07-20 | User portrait generation method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010695723.3A CN111738778B (en) | 2020-07-20 | 2020-07-20 | User portrait generation method and device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111738778A CN111738778A (en) | 2020-10-02 |
| CN111738778B true CN111738778B (en) | 2020-12-01 |
Family
ID=72654959
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010695723.3A Active CN111738778B (en) | 2020-07-20 | 2020-07-20 | User portrait generation method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111738778B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113052434A (en) * | 2021-02-26 | 2021-06-29 | 一智科技(成都)有限公司 | Construction object portrait labeling method and system |
| CN113987018A (en) * | 2021-10-27 | 2022-01-28 | 平安国际智慧城市科技股份有限公司 | Character feature mining method, device, equipment and storage medium |
| CN114819924B (en) * | 2022-06-28 | 2022-09-23 | 杭银消费金融股份有限公司 | Enterprise information push processing method and device based on portrait analysis |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009251717A (en) * | 2008-04-02 | 2009-10-29 | Yahoo Japan Corp | Advertisement distribution device, system, and method |
| CN105608477A (en) * | 2016-03-01 | 2016-05-25 | 吕云 | Method and system for matching portraits with positions |
| CN106056407A (en) * | 2016-06-03 | 2016-10-26 | 北京网智天元科技股份有限公司 | Online banking user portrait drawing method and equipment based on user behavior analysis |
| CN106447490A (en) * | 2016-09-26 | 2017-02-22 | 广州速鸿信息科技有限公司 | Credit investigation application method based on user figures |
| CN108021673A (en) * | 2017-12-06 | 2018-05-11 | 北京拉勾科技有限公司 | A kind of user interest model generation method, position recommend method and computing device |
| CN109558530A (en) * | 2018-10-23 | 2019-04-02 | 深圳壹账通智能科技有限公司 | User's portrait automatic generation method and system based on data processing |
| CN109657039A (en) * | 2018-11-15 | 2019-04-19 | 中山大学 | A kind of track record information extraction method based on the double-deck BiLSTM-CRF |
| CN110209767A (en) * | 2019-05-28 | 2019-09-06 | 重庆大学 | A kind of user's portrait construction method |
| CN110348703A (en) * | 2019-06-21 | 2019-10-18 | 平安科技（深圳）有限公司 | Data processing method, device and electronic equipment based on user behavior portrait |
| CN110929021A (en) * | 2018-08-31 | 2020-03-27 | 阿里巴巴集团控股有限公司 | Text information generating method and text information generating device |
-
2020
- 2020-07-20 CN CN202010695723.3A patent/CN111738778B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009251717A (en) * | 2008-04-02 | 2009-10-29 | Yahoo Japan Corp | Advertisement distribution device, system, and method |
| CN105608477A (en) * | 2016-03-01 | 2016-05-25 | 吕云 | Method and system for matching portraits with positions |
| CN106056407A (en) * | 2016-06-03 | 2016-10-26 | 北京网智天元科技股份有限公司 | Online banking user portrait drawing method and equipment based on user behavior analysis |
| CN106447490A (en) * | 2016-09-26 | 2017-02-22 | 广州速鸿信息科技有限公司 | Credit investigation application method based on user figures |
| CN108021673A (en) * | 2017-12-06 | 2018-05-11 | 北京拉勾科技有限公司 | A kind of user interest model generation method, position recommend method and computing device |
| CN110929021A (en) * | 2018-08-31 | 2020-03-27 | 阿里巴巴集团控股有限公司 | Text information generating method and text information generating device |
| CN109558530A (en) * | 2018-10-23 | 2019-04-02 | 深圳壹账通智能科技有限公司 | User's portrait automatic generation method and system based on data processing |
| CN109657039A (en) * | 2018-11-15 | 2019-04-19 | 中山大学 | A kind of track record information extraction method based on the double-deck BiLSTM-CRF |
| CN110209767A (en) * | 2019-05-28 | 2019-09-06 | 重庆大学 | A kind of user's portrait construction method |
| CN110348703A (en) * | 2019-06-21 | 2019-10-18 | 平安科技（深圳）有限公司 | Data processing method, device and electronic equipment based on user behavior portrait |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111738778A (en) | 2020-10-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111738778B (en) | User portrait generation method and device, computer equipment and storage medium | |
| CN107368521B (en) | Knowledge recommendation method and system based on big data and deep learning | |
| CN111506723A (en) | Question-answer response method, device, equipment and storage medium | |
| CN112380359B (en) | Knowledge graph-based training resource allocation method, device, equipment and medium | |
| CN112328646B (en) | Multitask course recommendation method and device, computer equipment and storage medium | |
| CN117114514B (en) | Talent information analysis management method, system and device based on big data | |
| CN113657495A (en) | Insurance product recommendation method, device and equipment based on probability prediction model | |
| Ejohwomu et al. | A resilient approach to modelling the supply and demand of platelets in the United Kingdom blood supply chain | |
| CN115545516A (en) | Performance data processing method, device and system based on process engine | |
| CN112288337B (en) | Behavior recommendation method, behavior recommendation device, behavior recommendation equipment and behavior recommendation medium | |
| CN111639706A (en) | Personal risk portrait generation method based on image set and related equipment | |
| CN112818028B (en) | Data index screening method and device, computer equipment and storage medium | |
| CN114862140A (en) | Behavior analysis-based potential evaluation method, device, equipment and storage medium | |
| CN114756669A (en) | Intelligent analysis method and device for problem intention, electronic equipment and storage medium | |
| CN114003592A (en) | Loan risk assessment method based on artificial intelligence and related equipment | |
| CN113807103A (en) | Recruitment method, device, equipment and storage medium based on artificial intelligence | |
| CN113627797A (en) | Image generation method and device for employee enrollment, computer equipment and storage medium | |
| CN111651452A (en) | Data storage method and device, computer equipment and storage medium | |
| CN115775140B (en) | Urban talent planning and human resource intelligent allocation system and method | |
| CN112328752B (en) | Course recommendation method and device based on search content, computer equipment and medium | |
| CN115099680A (en) | Risk management method, device, equipment and storage medium | |
| CN114219663A (en) | Product recommendation method and device, computer equipment and storage medium | |
| CN114595321A (en) | Question marking method and device, electronic equipment and storage medium | |
| CN113987351A (en) | Artificial intelligence based intelligent recommendation method and device, electronic equipment and medium | |
| CN112365051A (en) | Agent retention prediction method and device, computer 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 |