CN111667306A - Customized production-oriented customer demand identification method, system and terminal - Google Patents
Customized production-oriented customer demand identification method, system and terminal Download PDFInfo
- Publication number
- CN111667306A CN111667306A CN202010461695.9A CN202010461695A CN111667306A CN 111667306 A CN111667306 A CN 111667306A CN 202010461695 A CN202010461695 A CN 202010461695A CN 111667306 A CN111667306 A CN 111667306A
- Authority
- CN
- China
- Prior art keywords
- customer
- word
- rpa
- information
- customer demand
- 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
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- 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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0621—Item configuration or customization
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- General Engineering & Computer Science (AREA)
- Finance (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Marketing (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Game Theory and Decision Science (AREA)
- Human Computer Interaction (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention belongs to the technical field of information processing, and relates to a customized production-oriented customer demand identification method, a customized production-oriented customer demand identification system and a customized production-oriented customer demand identification terminal; the demand identification method comprises the steps that an RPA management platform calls an RPA robot to collect customer demand information and stores the customer demand information into a customer demand information database; calling a database interface to receive customer demand information data; preprocessing the collected customer requirement information at least including stop words; performing word segmentation on the preprocessed data by adopting a word segmentation mode based on a word list; calculating a word vector of the text data after word segmentation and taking the word vector as a data feature; inputting the obtained data characteristics into a preset demand classification model, and identifying a classification result of the customer demand information; the invention can carry out a series of analysis preprocessing on the client customized demand information collected in batch, combines a deep learning model, carries out demand identification and classification on the client customized demand, and obtains accurate client demand with less interaction.
Description
Technical Field
The invention belongs to the technical field of information processing, and relates to a customized production-oriented customer demand identification method, a customized production-oriented customer demand identification system and a customized production-oriented customer demand identification terminal.
Background
With the global development of economy, the market competition is increasingly intense. Enterprises need to provide products meeting individual and diversified requirements of the customers in order to win the customers in fierce competition, and therefore customized production facing the customers becomes a key means for the enterprises to win the market. Customized production is directed to customer requirements, and enterprises are required to respond quickly to different requirements set by different customers. The requirements of various customers are acquired, identified and classified, so that accurate and real requirements are mapped to product functions and then mapped to product structures for realizing the product functions, and finally customized production is completed.
For different acquired requirements of different customers, the non-structural data of the customers and the structural data of enterprises need to be matched and classified, and the requirements of the customers on various aspects such as color, configuration, delivery address, delivery cycle and the like are accurately positioned by using natural language processing based on deep learning and a small amount of interaction, so that higher satisfaction is obtained.
Disclosure of Invention
In view of the above, the present invention provides a method, a system, and a terminal for identifying customer requirements for customized production, which establish a requirement classification model by analyzing collected customer requirements or creative transmission information and using methods of data cleaning, data mining, and feature extraction to meet the objective of customized production.
In order to achieve the purpose, the invention provides the following technical scheme:
in a first aspect of the invention, the invention provides a customized production-oriented customer demand identification method; the identification method comprises the following steps:
the RPA management platform calls an RPA robot to collect customer requirement information and stores the customer requirement information into a customer requirement information database;
calling a database interface to receive customer demand information data;
preprocessing the collected customer requirement information at least including stop words;
performing word segmentation on the preprocessed data by adopting a word segmentation mode based on a word list;
calculating a word vector of the text data after word segmentation and taking the word vector as a data feature;
and inputting the obtained data characteristics into a preset requirement classification model, and identifying the classification result of the customer requirement information.
Further, performing word segmentation on the preprocessed data in a jieba word segmentation mode, which specifically comprises:
constructing a Trie tree word segmentation model, collecting preprocessed customer demand information sentences, and cleaning the customer demand information sentences;
constructing a prefix dictionary for the cleaned sentences by utilizing the Trie tree word segmentation model;
scanning based on the prefix dictionary to generate all possible word forming conditions of each Chinese character in the information sentence required by the customer;
segmenting input customer requirement information sentences to obtain directed acyclic graphs formed by all possible segmentations;
and calculating the maximum probability path in the directed acyclic graph formed by all possible segmentation modes through a dynamic programming algorithm, thereby obtaining a final segmentation mode, namely a word segmentation result.
Further, a word vector of the customer requirement information is calculated by adopting a TF-IDF text feature extraction method, and the method specifically comprises the following steps:
extracting TF characteristics in a customer requirement information text;
extracting IDF characteristics in a customer demand information text;
superposing the TF characteristics and the IDF characteristics, and extracting TF-IDF characteristics in the customer requirement information text;
the calculation formula of the TF characteristics is expressed as;
in the formula, the numerator is the frequency of the characteristic word t appearing in the text, and the denominator is the number of all the characteristic words in the text; the calculated result is the word frequency of a certain characteristic word;
the calculation formula of the IDF characteristic is expressed as;
where | D | represents the total number of texts in the corpus, | D |tiL represents the number of the characteristic words ti contained in the text; to prevent this word from not existing in the corpus, i.e., denominator 0, 1+ | D is usedtiTaking | as a denominator;
the calculation formula of the TF-IDF characteristics is expressed as follows;
TF-IDF=TF×IDF。
further, a TextCNN model is adopted as a preset requirement classification model, and the operation process specifically includes:
using a word vector trained in advance as an embedding layer, and inputting processed data characteristics to obtain an embedding expression;
after the embedded representation is obtained, the embedded representation is further input into a convolution layer, n-gram characteristics of a customer for product requirement individualized customization including color, configuration, receiving address and lead time are extracted through convolution, and a corresponding convolution kernel is activated;
inputting the convolution extracted n-gram features into a maximum pooling layer, and extracting features with the maximum activation degree;
inputting the features with the maximum activation degree into the full connection layer, and outputting the demand classification of different customers for the customized products.
In a second aspect of the present invention, the present invention further provides a customer demand identification system for customized production, the system comprising:
the RPA management platform is used for deploying and executing RPA robots of different acquisition processes for different demand processes and linking the RPA robots to the RPA management platform;
the RPA robot is used for acquiring the client requirement information in batches and uploading the client requirement information to the RPA management platform;
the customer demand information database is used for storing customer demand information in the RPA management platform by a user;
the database interface module is used for calling the customer requirement information in the customer requirement information database;
the preprocessing module is used for preprocessing the collected customer demand information at least including stop words; (ii) a
The word segmentation module is used for segmenting words of the preprocessed data in a word segmentation mode based on a word list;
the feature extraction module is used for extracting word vectors of the text data after word segmentation and taking the word vectors as data features;
and the classification module is used for inputting the obtained data characteristics into a preset demand classification model and identifying the classification result of the customer demand information.
In a third aspect of the present invention, the present invention provides a customized production oriented customer requirement identification terminal, which is characterized by comprising a processor and a memory, wherein the memory stores a computer program capable of running on the processor, and the processor executes the program to realize the customized production oriented customer requirement identification method.
The invention has the beneficial effects that:
according to the invention, the RPA robot can be used for collecting the customer requirement information in batch, the customized production data of the customer can be well utilized from the customer requirement information database, the data is subjected to a series of analysis preprocessing, the deep learning model TextCNN is combined, the customized requirements of the customer are subjected to requirement identification and classification, and the accurate customer requirements are obtained by small interaction.
Drawings
In order to make the object, technical scheme and beneficial effect of the invention more clear, the invention provides the following drawings for explanation:
FIG. 1 is a schematic diagram of a customized production oriented customer requirement identification method of the present invention;
FIG. 2 is a flow chart of a method for constructing a customer requirement information database according to the present invention;
FIG. 3 is a flow chart of a jieba word segmentation method employed by the present invention;
FIG. 4 is a flow chart of a requirement classification result of a customer classification by a preset requirement classification model according to the present invention;
FIG. 5 is a schematic diagram of a customized production oriented customer requirement identification system according to the present invention;
fig. 6 is a schematic structural diagram of a customized production oriented customer demand identification terminal in the invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
It should be noted that the execution subject of the embodiment of the present invention may be various types of terminals, and the terminal may be, for example, a computer, a server, a tablet computer, a Personal Digital Assistant (PDA), a Mobile Internet Device (MID), and other devices capable of performing text processing, which is not limited in this respect.
Customer needs refer to the actual needs of the customer being understood extensively and deeply to help the enterprise make the right decisions. Whether the economy is low or high, the survival development of enterprises should always be guided by the customer requirements, and the development direction of the business is continuously improved only by the customer requirements, so that the favor of more consumers can be won, and the customer satisfaction is improved.
Since the collected customer requirement information is generally unstructured data, in order to match and classify the unstructured data and the structural data of the enterprise, the customer requirement information of the customized production needs to be identified and classified, as shown in fig. 1, which is a schematic diagram of the requirement identification method of the present invention. FIG. 1 depicts the basic flow of the process of the present invention.
101. The RPA management platform calls an RPA robot to collect customer requirement information and stores the customer requirement information into a customer requirement information database;
as shown in fig. 2, the customer demand information database may be constructed in the following manner;
1011. the RPA management platform deploys and executes RPA robots with different acquisition processes for different required processes, and links the RPA robots to the RPA management platform;
1012. a client sends a request for submitting a demand to an RPA management platform, the RPA management platform automatically accepts the request submitted by the client, and determines the time and the number for calling the RPA robot based on the priority and the information quantity of the request submitted;
1013. distributing tasks required to be executed by the request to one or more RPA robots at the calling time;
1014. after the RPA robot is called, the execution workflow collects corresponding customer demand information and transmits the demand information back to the RPA management platform;
1015. and the RPA management platform stores the acquired customer demand information into a database.
The client demand information in the invention belongs to customized production data of clients, for example, the client sends demand information that the client needs toothpaste with a price of about 20 yuan and a whitening effect; the RPA robot acquires the demand information sent by the user, transmits the demand information back to the RPA management platform and then outputs the demand information to the customer demand information database through the data transmission interface.
102. Calling a database interface to receive customer demand information data;
at the moment, an interface of a customer demand information database is called, and a large amount of customer demand information is received; at this time, a rough classification may be performed according to the difference between the RPA robots, for example, the customer demand information collected by the first RPA robot and the customer demand information collected by the second RPA robot are respectively processed, and it is assumed that the first RPA robot collects demand information from a convenience store platform, for example, "i want a bottle of lemon-flavored soda water"; the second RPA robot collects the demand information from a certain household appliance sale platform, such as 'I want a 55-inch frameless TV set'; because the difference between the two kinds of requirement information is large, the data can be roughly classified according to the source of the data collected by the RPA robot when being processed.
103. Preprocessing the collected customer requirement information at least including stop words;
the stop words are words without practical significance, which can affect the normal semantic analysis of the text data.
104. Performing word segmentation on the preprocessed data by adopting a word segmentation mode based on a word list;
in the embodiment, unlike english which uses a space as a natural separator, when a chinese character is recognized semantically, a plurality of characters need to be combined into a word to express a true meaning, and thus, word segmentation processing needs to be performed on the text. The invention adopts the jieba word segmentation method as shown in the figure for preprocessing, as shown in figure 3, comprising:
1041. constructing a Trie tree word segmentation model, collecting preprocessed customer demand information sentences, and cleaning the customer demand information sentences;
1042. constructing a prefix dictionary for the cleaned sentences by utilizing the Trie tree word segmentation model;
1043. scanning based on the prefix dictionary to generate all possible word forming conditions of each Chinese character in the information sentence required by the customer;
1044. segmenting input customer requirement information sentences to obtain directed acyclic graphs formed by all possible segmentations;
1045. and calculating the maximum probability path in the directed acyclic graph formed by all possible segmentation modes through a dynamic programming algorithm, thereby obtaining a final segmentation mode, namely a word segmentation result.
Specifically, the input sentence and the word segmentation result are taken as two sequences, the sentence is an observation sequence, the word segmentation result is a state sequence, and the word segmentation result is obtained when the labeling of the state sequence is completed. Taking "white lemon flavored toothpaste" as an example, we know that the word segmentation result of "white fruity toothpaste" is "white/lemon flavored/toothpaste". For the participle state, we compute with 4-tag since 4-tag is used in the jieba participle. 4-tag, i.e. 4 possible states in the word for each word, B, M, E, S, respectively indicate Begin (the word is at the beginning of the word), Middle (the word is at the Middle of the word), End (the word is at the End of the word), and Single (the word is a Single word). The method comprises the following specific steps: "white/lemon flavor/toothpaste" are words combined from multiple characters, thus "white" and "color" are located in "white" B and E; "lemon", and "taste" are located in sequence at B, M, E of "lemon taste"; "teeth", "paste" are located at B and E of "toothpaste".
In some feasible embodiments, in the process of segmenting the client requirement information, the method and the device can improve the accuracy rate of segmenting words by technologies such as an external dictionary or new word discovery and the like, so as to assist the keyword extraction task.
For example, when the word segmentation processing of the customer requirement information text is assisted by the external dictionary based technology, the process of constructing the external dictionary may be: firstly, a word stock in the application field of the product and all word stocks in shopping software are collected through network search to obtain initial field words or phrases, then the initial field words or phrases can be screened, repeated words are filtered, and words with the word length being the preset length are reserved. Further, the word segmentation tool can be used for further screening, and unrealistic words such as conjunctions, prepositions, auxiliary verbs and the like are filtered out, so that the external dictionary of the corresponding field of the customer demand information text is obtained.
In a supplementary embodiment, when the customer requirement information belongs to long-section requirement information, for example, "i need one 200g, about thirty yuan, lemon-flavored Yunnan white drug powder toothpaste"; the method can also be used for preprocessing a client demand information text such as sentence segmentation, word segmentation, part-of-speech tagging, part-of-speech filtering and the like to obtain a word set D '{ D1, D2... Dn } of the client demand information text with words segmented, analyzing the client demand information text to obtain words corresponding to keywords of the client demand information text in the client demand information text, and generating a keyword candidate set W' { W1, W2.. Wm }.
The Word set and the keyword candidate set of the customer requirement information text are embedded into Word-level vector space representation through Word embedding (for example, based on a Word2vec Word vector representation model, part-of-speech characteristics are comprehensively considered, a characteristic extraction mode is optimized, more effective words with representative characteristics are obtained), cosine similarity (D ', W ') between words in D ' and W ' is obtained, and cosine similarity (γ) between words in D ' can also be obtained.
Using the sum γ, an MMR (maximum boundary correlation) value is calculated. And finally, extracting Top-K keywords with the highest scores as the keywords of the section of the client requirement information text according to the MMR value, wherein the extracted keywords can be 200g, thirty yuan, lemon flavor, Yunnan white drug powder and toothpaste.
105. Calculating a word vector of the text data after word segmentation and taking the word vector as a data feature;
because the client demand information after word segmentation is acquired, word vectors of the text data after word segmentation can be calculated by adopting a method based on statistics (co-occurrence matrix and SVD decomposition) or a method based on neural networks with different structures such as word2vec and the like.
In a preferred embodiment, the method for calculating the word vector of the customer requirement information by adopting the TF-IDF text feature extraction method specifically comprises the following steps:
extracting TF characteristics in a customer requirement information text;
extracting IDF characteristics in a customer demand information text;
superposing the TF characteristics and the IDF characteristics, and extracting TF-IDF characteristics in the customer requirement information text;
the calculation formula of the TF characteristics is expressed as;
in the formula, ni,jMeaning the word tiIn the text dj∑knk,jIs shown in the text djAll the words in the Chinese; tf isi,jI.e. the word tiI.e. the characteristic word tiThe word frequency of;
the calculation formula of the IDF characteristic is expressed as;
where | D | represents the total number of texts in the corpus, | D |tiI represents that the text contains a characteristic word tiThe number of (2); to prevent this word from not existing in the corpus, i.e., denominator 0, 1+ | D is usedtiTaking | as a denominator;
the calculation formula of the TF-IDF characteristics is expressed as follows;
TF-IDF=TF×IDF。
106. and inputting the obtained data characteristics into a preset requirement classification model, and identifying the classification result of the customer requirement information.
Specifically, reference is made to fig. 4:
1061. using a word vector trained in advance as an embedding layer, and inputting processed data characteristics to obtain an embedding expression;
1062. after the embedded representation is obtained, the embedded representation is further input into a convolution layer, n-gram characteristics of a customer for product requirement individualized customization including color, configuration, receiving address and lead time are extracted through convolution, and a corresponding convolution kernel is activated;
1063. inputting the convolution extracted n-gram features into a maximum pooling layer, and extracting features with the maximum activation degree;
1064. inputting the features with the maximum activation degree into the full connection layer, and outputting the demand classification of different customers for the customized products.
The following are embodiments of the apparatus of the present invention, which are used to implement the method of the first embodiment and the method of the second embodiment of the present invention, and for convenience of description, only relevant portions of the embodiments of the present invention, and specifically, portions not disclosed, are shown.
Referring to FIG. 5, in some possible embodiments, a customized production oriented customer demand identification system is shown in FIG. 5, the system comprising:
the RPA management platform is used for deploying and executing RPA robots of different acquisition processes for different demand processes and linking the RPA robots to the RPA management platform;
the RPA robot is used for acquiring the client requirement information in batches and uploading the client requirement information to the RPA management platform;
the customer demand information database is used for storing customer demand information in the RPA management platform by a user;
the database interface module is used for calling the customer requirement information in the customer requirement information database;
the preprocessing module is used for preprocessing the collected customer demand information at least including stop words; (ii) a
The word segmentation module is used for segmenting words of the preprocessed data in a word segmentation mode based on a word list;
the feature extraction module is used for extracting word vectors of the text data after word segmentation and taking the word vectors as data features;
and the classification module is used for inputting the obtained data characteristics into a preset demand classification model and identifying the classification result of the customer demand information.
The embodiment of the invention also provides a customized production-oriented customer demand identification terminal, which comprises a processor and a memory, wherein the memory is stored with a computer program capable of running on the processor, and the processor executes the program to realize a customized production-oriented customer demand identification method.
As shown in fig. 6, the terminal in the embodiment of the present invention includes: at least one input device; at least one processor, such as a CPU; at least one memory; at least one output device, the input device, the processor, the memory, and the output device being connected by a bus. Wherein the bus is used for enabling connection communication between these components. The input device and the output device of the apparatus in the embodiment of the present invention may be wired transmission ports, or may also be wireless devices, for example, including an antenna apparatus, which is used for performing signaling or data communication with other node devices.
The processor may be a Central Processing Unit (CPU), a Network Processor (NP), or a combination of a CPU and an NP.
The processor may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a Programmable Logic Device (PLD), or a combination thereof. The PLD may be a Complex Programmable Logic Device (CPLD), a field-programmable gate array (FPGA), a General Array Logic (GAL), or any combination thereof.
The memory may include volatile memory (volatile memory), such as random-access memory (RAM); the memory may also include a non-volatile memory (non-volatile) such as a flash memory (flash memory), a Hard Disk Drive (HDD) or a solid-state drive (SSD); the memory may also comprise a combination of memories of the kind described above.
Optionally, the memory is also used for storing program instructions. The processor may call the program instructions stored in the memory to implement the methods according to the first and second embodiments of the present invention.
The bus may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.
Specifically, the processor 2000 is configured to input a customer requirement information text; preprocessing the collected customer requirement information at least including stop words; performing word segmentation on the preprocessed data by adopting a word segmentation mode based on a word list; calculating a word vector of the text data after word segmentation and taking the word vector as a data feature; and inputting the obtained data characteristics into a preset requirement classification model, and identifying the classification result of the customer requirement information. In the embodiments shown in fig. 1 to fig. 4, the method flows of the steps may be implemented based on the structure of the terminal.
Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments may be implemented by associated hardware instructed by a program, which may be stored in a computer-readable storage medium, and the storage medium may include: ROM, RAM, magnetic or optical disks, and the like.
The above-mentioned embodiments, which further illustrate the objects, technical solutions and advantages of the present invention, should be understood that the above-mentioned embodiments are only preferred embodiments of the present invention, and should not be construed as limiting the present invention, and any modifications, equivalents, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (7)
1. A customer demand identification method facing customized production is characterized in that,
the RPA management platform calls an RPA robot to collect customer requirement information and stores the customer requirement information into a customer requirement information database;
calling a database interface to receive customer demand information data;
preprocessing the collected customer requirement information at least including stop words;
performing word segmentation on the preprocessed data by adopting a word segmentation mode based on a word list;
calculating a word vector of the text data after word segmentation and taking the word vector as a data feature;
and inputting the obtained data characteristics into a preset requirement classification model, and identifying the classification result of the customer requirement information.
2. The method for identifying customer demand oriented to customized production according to claim 1, wherein the step of the RPA management platform invoking the RPA robot to collect customer demand information and storing the customer demand information into the customer demand information database comprises:
the RPA management platform deploys and executes RPA robots with different acquisition processes for different required processes, and links the RPA robots to the RPA management platform;
a client sends a request for submitting a demand to an RPA management platform, the RPA management platform automatically accepts the request submitted by the client, and determines the time and the number for calling the RPA robot based on the priority and the information quantity of the request submitted;
distributing tasks required to be executed by the request to one or more RPA robots at the calling time;
after the RPA robot is called, the execution workflow collects corresponding customer demand information and transmits the demand information back to the RPA management platform;
and the RPA management platform stores the acquired customer demand information into a database.
3. The method for identifying customer demands for customized production according to claim 1, wherein the participling is performed on the preprocessed data in a jieba participling manner, and specifically comprises:
constructing a Trie tree word segmentation model, collecting preprocessed customer demand information sentences, and cleaning the customer demand information sentences;
constructing a prefix dictionary for the cleaned sentences by utilizing the Trie tree word segmentation model;
scanning based on the prefix dictionary to generate all possible word forming conditions of each Chinese character in the information sentence required by the customer;
segmenting input customer requirement information sentences to obtain directed acyclic graphs formed by all possible segmentations;
and calculating the maximum probability path in the directed acyclic graph formed by all possible segmentation modes through a dynamic programming algorithm, thereby obtaining a final segmentation mode, namely a word segmentation result.
4. The customer demand identification method facing customized production according to claim 1, wherein the calculating of the word vector of the customer demand information by using the TF-IDF text feature extraction method specifically comprises:
extracting TF characteristics in a customer requirement information text;
extracting IDF characteristics in a customer demand information text;
superposing the TF characteristics and the IDF characteristics, and extracting TF-IDF characteristics in the customer requirement information text;
the calculation formula of the TF characteristics is expressed as;
in the formula, ni,jMeaning the word tiIn the text dj∑knk,jIs shown in the text djAll the words in the Chinese; tf isi,jI.e. the word tiI.e. the characteristic word tiThe word frequency of;
the calculation formula of the IDF characteristic is expressed as;
where | D | represents the total number of texts in the corpus, | D |tiI represents that the text contains a characteristic word tiThe number of (2);
the calculation formula of the TF-IDF characteristics is expressed as follows;
TF-IDF=TF×IDF。
5. the method for identifying customer demand for customized production according to claim 1, wherein a TextCNN model is adopted as a preset demand classification model, and the operation process specifically includes:
using a word vector trained in advance as an embedding layer, and inputting processed data characteristics to obtain an embedding expression;
after the embedded representation is obtained, the embedded representation is further input into a convolution layer, n-gram characteristics of a customer for product requirement individualized customization including color, configuration, receiving address and lead time are extracted through convolution, and a corresponding convolution kernel is activated;
inputting the convolution extracted n-gram features into a maximum pooling layer, and extracting features with the maximum activation degree;
inputting the features with the maximum activation degree into the full connection layer, and outputting the demand classification of different customers for the customized products.
6. A customized production oriented customer demand identification system, the system comprising:
the RPA management platform is used for deploying and executing RPA robots of different acquisition processes for different demand processes and linking the RPA robots to the RPA management platform;
the RPA robot is used for acquiring the client requirement information in batches and uploading the client requirement information to the RPA management platform;
the customer demand information database is used for storing customer demand information in the RPA management platform by a user;
the database interface module is used for calling the customer requirement information in the customer requirement information database;
the preprocessing module is used for preprocessing the collected customer demand information at least including stop words; (ii) a
The word segmentation module is used for segmenting words of the preprocessed data in a word segmentation mode based on a word list;
the feature extraction module is used for extracting word vectors of the text data after word segmentation and taking the word vectors as data features;
and the classification module is used for inputting the obtained data characteristics into a preset demand classification model and identifying the classification result of the customer demand information.
7. A customized production oriented customer demand identification terminal comprising a processor and a memory, the memory having stored thereon a computer program operable on the processor, the processor when executing the program implementing the method according to any of claims 1 to 5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010461695.9A CN111667306A (en) | 2020-05-27 | 2020-05-27 | Customized production-oriented customer demand identification method, system and terminal |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010461695.9A CN111667306A (en) | 2020-05-27 | 2020-05-27 | Customized production-oriented customer demand identification method, system and terminal |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111667306A true CN111667306A (en) | 2020-09-15 |
Family
ID=72384737
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010461695.9A Pending CN111667306A (en) | 2020-05-27 | 2020-05-27 | Customized production-oriented customer demand identification method, system and terminal |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111667306A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112256770A (en) * | 2020-11-13 | 2021-01-22 | 国网江苏省电力有限公司南通供电分公司 | Intelligent data analysis platform based on intelligent data acquisition and analysis |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106919673A (en) * | 2017-02-21 | 2017-07-04 | 浙江工商大学 | Text mood analysis system based on deep learning |
CN110069252A (en) * | 2019-04-11 | 2019-07-30 | 浙江网新恒天软件有限公司 | A kind of source code file multi-service label mechanized classification method |
CN110516074A (en) * | 2019-10-23 | 2019-11-29 | 中国人民解放军国防科技大学 | Website theme classification method and device based on deep learning |
CN110689223A (en) * | 2019-08-22 | 2020-01-14 | 深圳市全通数码科技有限公司 | Robot batch management system based on RPA control |
CN111046175A (en) * | 2019-11-18 | 2020-04-21 | 杭州天翼智慧城市科技有限公司 | Self-learning-based electronic file classification method and device |
CN111159349A (en) * | 2019-12-30 | 2020-05-15 | 佰聆数据股份有限公司 | Automatic classification method, system, storage medium and computer equipment for power demand work orders based on natural language processing technology |
CN111159389A (en) * | 2019-12-31 | 2020-05-15 | 重庆邮电大学 | Keyword extraction method based on patent elements, terminal and readable storage medium |
CN111191030A (en) * | 2019-12-20 | 2020-05-22 | 北京淇瑀信息科技有限公司 | Single sentence intention identification method, device and system based on classification |
-
2020
- 2020-05-27 CN CN202010461695.9A patent/CN111667306A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106919673A (en) * | 2017-02-21 | 2017-07-04 | 浙江工商大学 | Text mood analysis system based on deep learning |
CN110069252A (en) * | 2019-04-11 | 2019-07-30 | 浙江网新恒天软件有限公司 | A kind of source code file multi-service label mechanized classification method |
CN110689223A (en) * | 2019-08-22 | 2020-01-14 | 深圳市全通数码科技有限公司 | Robot batch management system based on RPA control |
CN110516074A (en) * | 2019-10-23 | 2019-11-29 | 中国人民解放军国防科技大学 | Website theme classification method and device based on deep learning |
CN111046175A (en) * | 2019-11-18 | 2020-04-21 | 杭州天翼智慧城市科技有限公司 | Self-learning-based electronic file classification method and device |
CN111191030A (en) * | 2019-12-20 | 2020-05-22 | 北京淇瑀信息科技有限公司 | Single sentence intention identification method, device and system based on classification |
CN111159349A (en) * | 2019-12-30 | 2020-05-15 | 佰聆数据股份有限公司 | Automatic classification method, system, storage medium and computer equipment for power demand work orders based on natural language processing technology |
CN111159389A (en) * | 2019-12-31 | 2020-05-15 | 重庆邮电大学 | Keyword extraction method based on patent elements, terminal and readable storage medium |
Non-Patent Citations (1)
Title |
---|
黎巎: "《旅游大数据研究》", vol. 1, 31 July 2018, 中国经济出版社, pages: 55 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112256770A (en) * | 2020-11-13 | 2021-01-22 | 国网江苏省电力有限公司南通供电分公司 | Intelligent data analysis platform based on intelligent data acquisition and analysis |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Mai et al. | Joint sentence and aspect-level sentiment analysis of product comments | |
CN107705066B (en) | Information input method and electronic equipment during commodity warehousing | |
CN109992646B (en) | Text label extraction method and device | |
CN108304439B (en) | Semantic model optimization method and device, intelligent device and storage medium | |
CN112131350B (en) | Text label determining method, device, terminal and readable storage medium | |
CN111680159B (en) | Data processing method and device and electronic equipment | |
CN110909164A (en) | Text enhancement semantic classification method and system based on convolutional neural network | |
US10853697B2 (en) | System and method for monitoring online retail platform using artificial intelligence and fixing malfunction | |
CN110298029B (en) | Friend recommendation method, device, equipment and medium based on user corpus | |
CN112085565B (en) | Deep learning-based information recommendation method, device, equipment and storage medium | |
CN109918560A (en) | A kind of answering method and device based on search engine | |
KR20200007969A (en) | Information processing methods, terminals, and computer storage media | |
CN109086265B (en) | Semantic training method and multi-semantic word disambiguation method in short text | |
US20170109633A1 (en) | Comment-comment and comment-document analysis of documents | |
WO2017198031A1 (en) | Semantic parsing method and apparatus | |
CN112989208B (en) | Information recommendation method and device, electronic equipment and storage medium | |
CN112559684A (en) | Keyword extraction and information retrieval method | |
CN110287314B (en) | Long text reliability assessment method and system based on unsupervised clustering | |
CN112926308B (en) | Method, device, equipment, storage medium and program product for matching text | |
CN111309916B (en) | Digest extracting method and apparatus, storage medium, and electronic apparatus | |
CN112330455A (en) | Method, device, equipment and storage medium for pushing information | |
WO2016036345A1 (en) | External resource identification | |
CN112182145A (en) | Text similarity determination method, device, equipment and storage medium | |
CN110990563A (en) | Artificial intelligence-based traditional culture material library construction method and system | |
CN112580332A (en) | Enterprise portrait method based on label layering and deepening modeling |
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 |