CN113298559A - Commodity applicable crowd recommendation method, system, device and storage medium - Google Patents
Commodity applicable crowd recommendation method, system, device and storage medium Download PDFInfo
- Publication number
- CN113298559A CN113298559A CN202110535714.2A CN202110535714A CN113298559A CN 113298559 A CN113298559 A CN 113298559A CN 202110535714 A CN202110535714 A CN 202110535714A CN 113298559 A CN113298559 A CN 113298559A
- Authority
- CN
- China
- Prior art keywords
- commodity
- label
- description
- labels
- applicable
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0269—Targeted advertisements based on user profile or attribute
- G06Q30/0271—Personalized advertisement
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/242—Query formulation
- G06F16/243—Natural language query formulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2457—Query processing with adaptation to user needs
- G06F16/24573—Query processing with adaptation to user needs using data annotations, e.g. user-defined metadata
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/285—Clustering or classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/237—Lexical tools
- G06F40/242—Dictionaries
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Strategic Management (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Library & Information Science (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a method, a system, a device and a storage medium for recommending commodity applicable people, wherein the method comprises the following steps: the method comprises the steps of obtaining commodity information, carrying out natural language processing on the commodity information to obtain a plurality of description labels, and establishing a first mapping relation between the description labels and commodities; constructing a training data set carrying description labels according to the first mapping relation, and training according to the training data set to obtain an applicable crowd classification model; the crowd classification model is used for generating a second mapping relation between a combination of a plurality of description labels and applicable crowd labels; acquiring a commodity to be marked, inputting an applicable crowd classification model, and giving an applicable crowd label to the commodity to be marked according to a second mapping relation; the method can more efficiently endow the commodity labels to determine the target consumer groups, is intelligent and automatic, can be flexibly applied to various commodities in various fields, further achieves the aim of reducing marketing cost, and can be widely applied to the technical field of pattern recognition.
Description
Technical Field
The invention relates to the technical field of pattern recognition, in particular to a method and a system for recommending commodity applicable people and a storage medium.
Background
In the process of selling goods, the demands of consumers are various, different consumers need to be promoted and publicized differently, and therefore, unnecessary human resources and waste of funds need to be reduced by determining target audiences. At present, a relatively wide method is to give a label of a target consumer group to a commodity before sale, and most of the current modes of giving the label of the target consumer group to the commodity adopt a manual labeling mode, so that a large amount of manpower and material resources are required to process massive commodity data, the efficiency is low, and meanwhile, the marketing cost is increased.
In the prior art, the information of products of cosmetics is manually processed, the product information description characteristics need to be manually identified, and then corresponding crowd labels are given. The product information is complex, the people label the product is lack of uniform and standard basis under the condition of multi-person operation, the manual processing speed is very slow, the product information processing requirement is faster and more accurate, and more products cannot be met; therefore, a method capable of automatically, intelligently and multi-dimensionally giving an applicable population label or a target consumer population label to a product is needed, and enterprises or manufacturers are helped to differentially and accurately locate a target population of the product.
Disclosure of Invention
In view of the above, to at least partially solve one of the above technical problems, an embodiment of the present invention is directed to an automated, intelligent, multi-dimensional recommendation method for commodity-suitable people, and a system, an apparatus and a computer-readable storage medium capable of implementing the method.
In a first aspect, the technical scheme of the invention provides a recommendation method for people suitable for commodities, which comprises the following steps:
acquiring commodity information, carrying out natural language processing on the commodity information to obtain a plurality of description labels, and establishing a first mapping relation between the description labels and commodities;
constructing a training data set carrying the description label according to the first mapping relation, and training according to the training data set to obtain an applicable crowd classification model; the crowd classification model is used for generating a second mapping relation between a combination of a plurality of the description labels and applicable crowd labels;
and acquiring the commodity to be marked, inputting the applicable crowd classification model, and endowing the commodity to be marked with the applicable crowd label according to the second mapping relation.
In some optional embodiments, the step of obtaining the commodity information, performing natural language processing on the commodity information to obtain a plurality of description tags, and establishing a first mapping relationship between the description tags and the commodities includes:
acquiring the commodity information from a primary warehouse, and performing word segmentation processing on the commodity information to obtain a plurality of description labels;
and performing similarity matching on the description label and the label in the secondary warehouse, and establishing a first mapping relation between the description label and the commodity according to a matching result.
In some optional embodiments, the step of obtaining the commodity information from a primary warehouse, performing word segmentation processing on the commodity information, and obtaining a plurality of the description tags includes:
loading a dictionary, and constructing a dictionary tree word segmentation model according to the dictionary;
cleaning the sentence text of the commodity information, and generating a directed acyclic graph according to the cleaned sentence text and the dictionary tree word segmentation model;
and determining a maximum segmentation combination according to the directed acyclic graph, and performing word segmentation on the maximum segmentation combination through a hidden Markov model and a Viterbi algorithm.
In some optional embodiments, the step of performing similarity matching between the description tag and a tag in a secondary warehouse, and establishing a first mapping relationship between the description tag and a commodity according to a matching result includes:
obtaining a first string describing a tag and a second string describing a tag in the secondary repository,
determining a maximum value of a character string length and a first parameter according to the first character string and the second character string, wherein the first parameter is the number of times of character replacement for converting the first character string into the second character string;
and determining the similarity according to the difference value between the first parameter and the maximum value of the length of the character string.
In some optional embodiments, the step of obtaining the to-be-marked product, inputting the applicable population classification model, and assigning the applicable population label to the to-be-marked product according to the second mapping relationship includes:
inputting the commodity information of the commodity to be marked into the applicable crowd classification model, and determining the conditional probability of the applicable crowd label according to the second mapping relation by the applicable crowd classification model;
and determining the maximum value in the conditional probability as the applicable crowd label of the commodity to be marked.
In a second aspect, the present invention further provides a recommendation system for people suitable for goods, including:
the natural language processing unit is used for acquiring commodity information, carrying out natural language processing on the commodity information to obtain a plurality of description labels, and establishing a first mapping relation between the description labels and commodities;
the model training unit is used for constructing a training data set carrying the description label according to the first mapping relation and training according to the training data set to obtain an applicable crowd classification model; the crowd classification model is used for generating a second mapping relation between a combination of a plurality of the description labels and applicable crowd labels;
and the commodity marking unit is used for acquiring the commodities to be marked, inputting the suitable crowd classification model and endowing the commodities to be marked with the suitable crowd labels according to the second mapping relation.
In a third aspect, the present invention provides a recommendation device for a commodity applicable crowd, including:
at least one processor;
at least one memory for storing at least one program;
when the at least one program is executed by the at least one processor, the at least one program causes the at least one processor to execute a method for recommending a commodity-suitable group of people according to the first aspect.
In a fourth aspect, the present invention also provides a storage medium, in which a processor-executable program is stored, and the processor-executable program is used for executing the method in the first aspect when being executed by a processor.
Advantages and benefits of the present invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention:
according to the technical scheme, the description labels of the commodities are obtained through natural language processing, the commodity information carrying the description labels is used as a training set to train an applicable crowd classification model, and the trained applicable crowd classification model can endow the applicable crowd labels to the commodities to be marked according to the mapping relation between the combination of the description labels and the applicable crowd labels; the method can more efficiently endow the commodity labels to determine the target consumer groups, and can be flexibly applied to various commodities in various fields in an intelligent and automatic processing mode, so that the aim of reducing the marketing cost is fulfilled.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a flowchart illustrating steps of a method for recommending people suitable for goods according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating steps for tag word segmentation according to an embodiment of the present invention;
fig. 3 is a flowchart illustrating steps of another method for recommending people suitable for goods according to an embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention. The step numbers in the following embodiments are provided only for convenience of illustration, the order between the steps is not limited at all, and the execution order of each step in the embodiments can be adapted according to the understanding of those skilled in the art.
In a first aspect, as shown in fig. 1, the technical solution of the present application provides an embodiment of a method for recommending people suitable for goods, wherein the method includes steps S100-S300:
s100, acquiring commodity information, performing natural language processing on the commodity information to obtain a plurality of description labels, and establishing a first mapping relation between the description labels and commodities;
the natural language processing includes, but is not limited to, performing processing such as sentence text cleaning, special character deletion, word segmentation and the like on text content of commodity information, the description tags obtained through the natural language processing are descriptive text characters such as the commodity name, components, efficacy and the like, the first mapping relation is the corresponding relation between the commodity and each description tag, and a certain type of commodity can be accurately determined from a primary warehouse according to a certain number of description tags. Taking cosmetics commodities as an example, the embodiment intelligently identifies the names, the components, the applicable skin types, the applicable ages and the effects of the cosmetics based on mass cosmetic data, and the identification algorithm adopts jieba word segmentation, stop word segmentation and Lewinstein short text similarity matching to establish a mapping relation data set of the products and each label.
S200, constructing a training data set carrying description labels according to the first mapping relation, and training according to the training data set to obtain an applicable crowd classification model;
the crowd classification model is used for generating a second mapping relation between a combination of a plurality of description labels and applicable crowd labels; specifically, through step S100, identifying description labels of the commodity information of the selected partial commodities, and integrating the commodity information carrying the description labels to obtain a training data set of the model; in the embodiment, a naive Bayes classification algorithm can be selected to classify the applicable people, and an applicable people classification model is constructed according to the principle of the naive Bayes classification algorithm. It should be understood that in the process of constructing the training data set, all applicable groups of the commodities of the type are determined first, and applicable group labels of all commodities in the training data set are given in advance. Exemplarily, taking a cosmetic commodity as an example, the embodiment firstly confirms the cosmetic description label and the applicable crowd label division, and carries out label classification to form a training sample, namely a training data set, wherein the quality of the applicable crowd classification model is largely determined by the attribute division of the cosmetic description label and the quality of the training sample; in the process of model training, the embodiment calculates the occurrence frequency of each preset applicable population label in a training sample and the conditional probability of each product label division on each applicable population label, records the result, and confirms that the model training is finished when the accuracy of the result meets a preset threshold value. It is understood that in the embodiment of the present application, the training method of the model may adopt a supervised learning manner, such as a naive bayes classification algorithm; and an unsupervised classification mode can be adopted to further reduce the work of a data preparation stage so as to improve the model training efficiency.
In addition, in the training process of the model, the model can be continuously trained and upgraded along with the description label and the updated training data set, and meanwhile, the development and training of multiple models are supported. In an embodiment, label x is described based on a training dataset1,x2,x3..xnThe training result is x with the matching relation between the applicable crowd labels C1,x2,x3..xnMatched with C under conditionsAnd the probability is maintained in the crowd label algorithm model, and the probability calculation formula is as follows:
s300, obtaining the commodity to be marked, inputting an applicable crowd classification model, and giving an applicable crowd label to the commodity to be marked according to a second mapping relation;
specifically, in this embodiment, the database storing the mass commodity information in step S100 is a primary warehouse, that is, the commodities in the primary warehouse are all known or determined to be suitable for the crowd label; the method comprises the steps of S100, performing word segmentation to finally obtain all description labels, storing the description labels in a secondary warehouse, in step S200, storing applicable population labels determined in a training data set in a tertiary warehouse, mapping specific commodities in the primary warehouse to obtain a mapping relation of a combination of a plurality of description labels in the secondary warehouse as a first mapping relation, and mapping the applicable population labels determined in the tertiary warehouse to obtain a mapping relation of the plurality of description labels in the secondary warehouse as a second mapping relation. According to the embodiment, the commodity information of the commodity to be marked is extracted after the commodity is obtained, the commodity information is input into the applicable crowd classification model after training, the applicable crowd label corresponding to the commodity is determined by identifying the description label and according to the second mapping relation of the model, and the applicable crowd label is stored in the three-level warehouse, namely the applicable crowd label library. And finally, matching the target population oriented product data based on the product applicable population label data in the three-level warehouse according to the existing user population and the applicable population label of the product, and then utilizing the existing PC intelligent recommendation system to popularize the target population oriented product such as App, small program, PC terminal and the like.
In some optional embodiments, the step S100 of acquiring the commodity information, performing natural language processing on the commodity information to obtain a plurality of description tags, and establishing a first mapping relationship between the description tags and the commodities may include the steps S110 and S120:
s110, acquiring commodity information from a primary warehouse, and performing word segmentation processing on the commodity information to obtain a plurality of description labels;
illustratively, taking a cosmetic product as an example, based on the product information of a certain cream in the primary warehouse, the description label is obtained by analyzing through a jieba word segmentation algorithm as follows:
skin-type labeling: all skin types
Efficacy labeling: moisturizing, moistening, pulling and tightening
Ingredient labeling: face cream
S120, similarity matching is carried out on the description label and a label in a preset secondary warehouse, and a first mapping relation between the description label and the commodity is established according to a matching result;
specifically, in the embodiment, according to step S110, an algorithm for identifying a label of a certain frost commodity, including but not limited to jieba word segmentation, stop word removal, reweistein short text matching, etc., is calculated to obtain: the product comprises efficacy labels of: moistening, moistening and moisturizing. And storing the association relationship between the commodities and the commodity description labels in a secondary warehouse.
In some optional embodiments, the step S110 of obtaining the commodity information from the primary warehouse, performing word segmentation processing on the commodity information, and obtaining a plurality of description labels may include steps S111 to S113:
s111, loading a dictionary, and constructing a dictionary tree word segmentation model according to the dictionary;
s112, cleaning the sentence text of the commodity information, and generating a directed acyclic graph according to the cleaned sentence text and the dictionary tree word segmentation model;
s113, determining a maximum segmentation combination according to the directed acyclic graph, and performing word segmentation on the maximum segmentation combination through a hidden Markov model and a Viterbi algorithm;
specifically, as shown in fig. 2, first, the embodiment loads the registration dictionary, and establishes a dictionary tree segmentation model (Trie tree segmentation model), where the Trie tree is used to count, sort, and store a large number of character strings (but not limited to character strings), and it uses the common prefixes of the character strings to reduce the query time, and reduce meaningless character string comparison to the maximum extent, and the query efficiency is higher than that of the hash tree. Meanwhile, sentence cleaning is carried out on the contents such as sentence text of commodity information, if the sentence contains special characters, the sentence is separated, the special characters are labeled as labels with unknown parts of speech, then a Directed Acyclic Graph (DAG) is established according to a Trie tree word segmentation model, a DAG data structure tracks the calculation and assignment of the median and the variable of a basic block, and the values from other places used in the block are represented as leaf nodes; operations on values are represented as internal nodes; the assignment of the new value is expressed as attaching the name of the target variable or the temporary variable to the node expressing the assignment; the advantage of building Trie numbers for DAG word segmentation is that space can be saved and fast search can be performed. Calculating a global probability Route (Route) according to the DAG graph to obtain a word frequency maximum segmentation combination based on a prefix dictionary; if the log-in words can be determined in the process, the marks are marked according to the dictionary. And secondly, respectively identifying the Chinese part and the English part through the token, respectively identifying the combination of English, number and time forms if an English and number dictionary combination mode exists, and giving corresponding labels. Then loading a hidden Markov Model (MHH) probability model diagram, and dynamically planning by a Viterbi (Viterbi) algorithm to obtain a word segmentation result.
In some possible embodiments, the step S120 of performing similarity matching between the description tag and the tag in the preset secondary warehouse, and establishing the first mapping relationship between the description tag and the commodity according to the matching result may include steps S121 to S123:
s121, acquiring a first character string describing a label and a second character string describing the label in a secondary warehouse;
s122, determining a maximum value of the length of the character string and a first parameter according to the first character string and the second character string;
s123, determining similarity according to the difference value between the first parameter and the maximum value of the length of the character string;
wherein, the first parameter is the number of times of character replacement for converting the first character string into the second character string; exemplarily, after the description label is obtained, the stop word is removed, and then the Lewenstein short text similarity matching algorithm is used for matching; the maximum value maxLen of the two comparison character string lengths is taken first, and the similarity is obtained by using 1- (operand/maxLen is needed). For example: the water supplement is performed with 0 operation, the maximum string length is 2, and the similarity 1- (0/2) is 1, namely 100% (similarity determination criterion: exceeding 60% is considered to be consistent). And obtaining the efficacy label of 'water supplement' of the cosmetic product information and the efficacy label 'water supplement' in the secondary warehouse according to the matching rate > 60%, successfully matching, and storing the matching relationship in the secondary warehouse.
In some embodiments, the step S400 of obtaining the to-be-marked commodity, inputting the applicable crowd classification model, and giving the to-be-marked commodity the applicable crowd label according to the second mapping relationship may further include the steps S410 to S420:
s410, inputting commodity information of the commodities to be marked into an applicable crowd classification model, and determining the conditional probability of the applicable crowd label according to the second mapping relation by the applicable crowd classification model;
s420, determining the maximum value in the conditional probability as an applicable crowd label of the commodity to be marked;
specifically, the embodiment uses the applicable population classification model to classify the commodities to be classified, the input of the model is the cosmetic products to be classified, the output of the model is the mapping relation between the cosmetic products to be classified and the population labels, the mapping relation is stored in the third-level warehouse, and based on the applicable population label data in the third-level warehouse, the embodiment matches the applicable population labels of the commodities with the existing user populations, the matching is successful, and then the accurate popularization is performed.
As shown in fig. 3, in an embodiment of the application, 10 pieces of commodity information are marked to train an applicable crowd classification model, and finally, the model analysis accuracy reaches over 90%. Meanwhile, the user at the user side and the user crowd labels are combined with the cosmetics crowd labels, interested cosmetics advertisements can be recommended to the user in a differentiated and accurate mode, the popularization accuracy of the cosmetics marketing advertisements and the product transaction rate are improved, the marketing input cost is reduced, and the achievement increase is promoted.
In a second aspect, the present application provides a recommendation system for a commodity suitability group for implementing the method of the first aspect, comprising:
the natural language processing unit is used for acquiring the commodity information, carrying out natural language processing on the commodity information to obtain a plurality of description labels, and establishing a first mapping relation between the description labels and the commodities;
the model training unit is used for constructing a training data set carrying description labels according to the first mapping relation and training according to the training data set to obtain an applicable crowd classification model; the crowd classification model is used for generating a second mapping relation between a combination of a plurality of description labels and applicable crowd labels;
and the commodity marking unit is used for acquiring the commodity to be marked, inputting the applicable crowd classification model and giving an applicable crowd label to the commodity to be marked according to the second mapping relation.
In a third aspect, the present disclosure further provides a recommendation device for a commodity applicable crowd, including at least one processor; at least one memory for storing at least one program; when the at least one program is executed by the at least one processor, the at least one program causes the at least one processor to execute a method for recommending commodity suitability for a population as in the first aspect.
An embodiment of the present invention further provides a storage medium storing a program, where the program is executed by a processor to implement the method in the first aspect.
From the above specific implementation process, it can be concluded that the technical solution provided by the present invention has the following advantages or advantages compared to the prior art:
the technical scheme of the application can give commodity labels more efficiently to determine target consumer groups, and intelligent and automatic processing modes can be flexibly applied to various commodities in various fields, so that the aim of reducing marketing cost is fulfilled.
In alternative embodiments, the functions/acts noted in the block diagrams may occur out of the order noted in the operational illustrations. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality/acts involved. Furthermore, the embodiments presented and described in the flow charts of the present invention are provided by way of example in order to provide a more thorough understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and in which sub-operations described as part of larger operations are performed independently.
Furthermore, although the present invention is described in the context of functional modules, it should be understood that, unless otherwise stated to the contrary, one or more of the functions and/or features may be integrated in a single physical device and/or software module, or one or more of the functions and/or features may be implemented in a separate physical device or software module. It will also be appreciated that a detailed discussion of the actual implementation of each module is not necessary for an understanding of the present invention. Rather, the actual implementation of the various functional modules in the apparatus disclosed herein will be understood within the ordinary skill of an engineer, given the nature, function, and internal relationship of the modules. Accordingly, those skilled in the art can, using ordinary skill, practice the invention as set forth in the claims without undue experimentation. It is also to be understood that the specific concepts disclosed are merely illustrative of and not intended to limit the scope of the invention, which is defined by the appended claims and their full scope of equivalents.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
While embodiments of the invention have been shown and described, it will be understood by those of ordinary skill in the art that: various changes, modifications, substitutions and alterations can be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.
While the preferred embodiments of the present invention have been illustrated and described, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (8)
1. A recommendation method for commodity applicable people is characterized by comprising the following steps:
acquiring commodity information, carrying out natural language processing on the commodity information to obtain a plurality of description labels, and establishing a first mapping relation between the description labels and commodities;
constructing a training data set carrying the description label according to the first mapping relation, and training according to the training data set to obtain an applicable crowd classification model; the crowd classification model is used for generating a second mapping relation between a combination of a plurality of the description labels and applicable crowd labels;
and acquiring the commodity to be marked, inputting the applicable crowd classification model, and endowing the commodity to be marked with the applicable crowd label according to the second mapping relation.
2. The method as claimed in claim 1, wherein the step of obtaining the commodity information, performing natural language processing on the commodity information to obtain a plurality of description tags, and establishing a first mapping relationship between the description tags and the commodities comprises:
acquiring the commodity information from a primary warehouse, and performing word segmentation processing on the commodity information to obtain a plurality of description labels;
and performing similarity matching on the description label and the label in the secondary warehouse, and establishing a first mapping relation between the description label and the commodity according to a matching result.
3. The method as claimed in claim 2, wherein the step of obtaining the commodity information from a primary warehouse, performing word segmentation processing on the commodity information to obtain a plurality of the description labels includes:
loading a dictionary, and constructing a dictionary tree word segmentation model according to the dictionary;
cleaning the sentence text of the commodity information, and generating a directed acyclic graph according to the cleaned sentence text and the dictionary tree word segmentation model;
and determining a maximum segmentation combination according to the directed acyclic graph, and performing word segmentation on the maximum segmentation combination through a hidden Markov model and a Viterbi algorithm.
4. The method as claimed in claim 2, wherein the step of matching the description label with the label in the secondary warehouse for similarity and establishing the first mapping relationship between the description label and the commodity according to the matching result comprises:
acquiring a first character string describing a label and a second character string describing the label in the secondary warehouse;
determining a maximum value of a character string length and a first parameter according to the first character string and the second character string, wherein the first parameter is the number of times of character replacement for converting the first character string into the second character string;
and determining the similarity according to the difference value between the first parameter and the maximum value of the length of the character string.
5. The method as claimed in claim 1, wherein the step of obtaining the product to be marked, inputting the model for classifying the applicable population, and assigning the label for the applicable population to the product to be marked according to the second mapping relationship comprises:
inputting the commodity information of the commodity to be marked into the applicable crowd classification model, and determining the conditional probability of the applicable crowd label according to the second mapping relation by the applicable crowd classification model;
and determining the maximum value in the conditional probability as the applicable crowd label of the commodity to be marked.
6. A system for recommending people suitable for goods, comprising:
the natural language processing unit is used for acquiring commodity information, carrying out natural language processing on the commodity information to obtain a plurality of description labels, and establishing a first mapping relation between the description labels and commodities;
the model training unit is used for constructing a training data set carrying the description label according to the first mapping relation and training according to the training data set to obtain an applicable crowd classification model; the crowd classification model is used for generating a second mapping relation between a combination of a plurality of the description labels and applicable crowd labels;
and the commodity marking unit is used for acquiring the commodities to be marked, inputting the suitable crowd classification model and endowing the commodities to be marked with the suitable crowd labels according to the second mapping relation.
7. A recommendation device for a commodity-suitable crowd, comprising:
at least one processor;
at least one memory for storing at least one program;
when executed by the at least one processor, cause the at least one processor to perform a method of recommending a commodity for use with a group of people as recited in any one of claims 1-5.
8. A storage medium having stored thereon a program executable by a processor, wherein the program executable by the processor is adapted to perform a method of recommending a commodity for a crowd according to any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110535714.2A CN113298559A (en) | 2021-05-17 | 2021-05-17 | Commodity applicable crowd recommendation method, system, device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110535714.2A CN113298559A (en) | 2021-05-17 | 2021-05-17 | Commodity applicable crowd recommendation method, system, device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113298559A true CN113298559A (en) | 2021-08-24 |
Family
ID=77322409
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110535714.2A Pending CN113298559A (en) | 2021-05-17 | 2021-05-17 | Commodity applicable crowd recommendation method, system, device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113298559A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114862329A (en) * | 2022-07-04 | 2022-08-05 | 江苏多飞网络科技有限公司 | Data storage method, access management method and management system for warehouse-in commodities |
| CN116720022A (en) * | 2023-08-07 | 2023-09-08 | 广州趣米网络科技有限公司 | Front-end page content display method and system based on SaaS system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120030060A1 (en) * | 2010-07-28 | 2012-02-02 | True Fit Corporation | Determining a likelihood of suitability based on historical data |
| CN110362670A (en) * | 2019-07-19 | 2019-10-22 | 中国联合网络通信集团有限公司 | Item property abstracting method and system |
| CN110428322A (en) * | 2019-06-12 | 2019-11-08 | 平安科技（深圳）有限公司 | A kind of adaptation method and device of business datum |
| CN110969222A (en) * | 2018-09-29 | 2020-04-07 | 西门子股份公司 | Information providing method and system |
| CN111737541A (en) * | 2020-06-30 | 2020-10-02 | 湖北亿咖通科技有限公司 | Semantic recognition and evaluation method supporting multiple languages |
-
2021
- 2021-05-17 CN CN202110535714.2A patent/CN113298559A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120030060A1 (en) * | 2010-07-28 | 2012-02-02 | True Fit Corporation | Determining a likelihood of suitability based on historical data |
| CN110969222A (en) * | 2018-09-29 | 2020-04-07 | 西门子股份公司 | Information providing method and system |
| CN110428322A (en) * | 2019-06-12 | 2019-11-08 | 平安科技（深圳）有限公司 | A kind of adaptation method and device of business datum |
| CN110362670A (en) * | 2019-07-19 | 2019-10-22 | 中国联合网络通信集团有限公司 | Item property abstracting method and system |
| CN111737541A (en) * | 2020-06-30 | 2020-10-02 | 湖北亿咖通科技有限公司 | Semantic recognition and evaluation method supporting multiple languages |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114862329A (en) * | 2022-07-04 | 2022-08-05 | 江苏多飞网络科技有限公司 | Data storage method, access management method and management system for warehouse-in commodities |
| CN114862329B (en) * | 2022-07-04 | 2022-09-20 | 江苏多飞网络科技有限公司 | Data storage method, access management method and management system for warehouse-in commodities |
| CN116720022A (en) * | 2023-08-07 | 2023-09-08 | 广州趣米网络科技有限公司 | Front-end page content display method and system based on SaaS system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110134757B (en) | Event argument role extraction method based on multi-head attention mechanism | |
| CN107633007B (en) | Commodity comment data tagging system and method based on hierarchical AP clustering | |
| CN107705066B (en) | Information input method and electronic equipment during commodity warehousing | |
| Gasmi et al. | LSTM recurrent neural networks for cybersecurity named entity recognition | |
| CN108255813B (en) | Text matching method based on word frequency-inverse document and CRF | |
| CN110705286A (en) | Comment information-based data processing method and device | |
| CN111104526A (en) | Financial label extraction method and system based on keyword semantics | |
| CN109522412B (en) | Text emotion analysis method, device and medium | |
| CN107832287A (en) | A kind of label identification method and device, storage medium, terminal | |
| CN111353306B (en) | Entity relationship and dependency Tree-LSTM-based combined event extraction method | |
| CN112069312B (en) | Text classification method based on entity recognition and electronic device | |
| CN113298559A (en) | Commodity applicable crowd recommendation method, system, device and storage medium | |
| CN113821605A (en) | Event extraction method | |
| CN113360647B (en) | 5G mobile service complaint source-tracing analysis method based on clustering | |
| CN111400449A (en) | Regular expression extraction method and device | |
| CN112862569B (en) | Product appearance style evaluation method and system based on image and text multi-modal data | |
| CN112685440B (en) | Structural query information expression method for marking search semantic role | |
| CN112685374B (en) | Log classification method and device and electronic equipment | |
| Manchanda et al. | Distant-supervised slot-filling for e-commerce queries | |
| CN110472032A (en) | More classification intelligent answer search methods of medical custom entities word part of speech label | |
| Joppi et al. | POP: mining POtential Performance of new fashion products via webly cross-modal query expansion | |
| CN115827871A (en) | Internet enterprise classification method, device and system | |
| CN110837735B (en) | Intelligent data analysis and identification method and system | |
| CN114372148A (en) | Data processing method based on knowledge graph technology and terminal equipment | |
| CN111339303A (en) | Text intention induction method and device based on clustering and automatic summarization |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210824 |