CN115393024A - Product data pushing method and device, computer equipment and storage medium - Google Patents
Product data pushing method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN115393024A CN115393024A CN202211173757.1A CN202211173757A CN115393024A CN 115393024 A CN115393024 A CN 115393024A CN 202211173757 A CN202211173757 A CN 202211173757A CN 115393024 A CN115393024 A CN 115393024A
- Authority
- CN
- China
- Prior art keywords
- product
- data
- session
- preset
- score
- 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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- 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
Landscapes
- Business, Economics & Management (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Engineering & Computer Science (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application relates to a natural language processing technology in the field of artificial intelligence, and provides a product data pushing method and device, computer equipment and a storage medium. The method comprises the following steps: the method comprises the steps of obtaining object session data, extracting session characteristic data of the object session data, constructing label data of a session object based on the session characteristic data, determining a product to be recommended according to the session characteristic data and a preset product keyword mapping table, pushing the product to be recommended and the label data, receiving product feedback data aiming at the product to be recommended and the label data, and adjusting the preset product keyword mapping table according to the product feedback data. By adopting the method, the intelligent recommendation of the product and the intelligent addition of the label data are realized, the tedious operation of manually maintaining the label by service personnel is omitted, and the error probability of manually adding the label is reduced. And the product keyword mapping table is adjusted according to the received product feedback data, so that the product recommendation accuracy can be improved, and refined product recommendation can be realized.
Description
Technical Field
The present application relates to the field of artificial intelligence natural language processing technologies, and in particular, to a method and an apparatus for pushing product data, a computer device, a storage medium, and a computer program product.
Background
In recent years, with the continuous development and rapid improvement of the internet, the fine operation of the electronic commerce is called as an important means for maintaining the amount of customers and guaranteeing the continuous increase of brands. The refined operation in the e-commerce field refers to matching different services and contents for users with different requirements through user grouping, so that the personalized requirements of the users are met.
At present, in the field of e-commerce, professional office management tools are important channels for enterprises to connect customers and communicate with the customers. In order to remember the preferences, characteristics and requirements of each client and to implement accurate recommendation of products to the client, a servicer or a servicer often marks the preferences, characteristics and requirements of each client by manually adding a label. However, when a clerk faces hundreds of customers, manually maintaining the labels requires a lot of time and labor, and is prone to errors, affecting the accuracy of the product recommendation.
Therefore, it is desirable to provide a solution that can improve the accuracy of product recommendation.
Disclosure of Invention
In view of the above, it is necessary to provide a product data pushing method, apparatus, computer device, computer readable storage medium and computer program product capable of improving the accuracy of product recommendation in view of the above technical problems.
In a first aspect, the present application provides a product data pushing method. The method comprises the following steps:
a product data pushing method, characterized in that the method comprises:
acquiring object session data;
extracting session characteristic data of the object session data, and constructing label data of the session object based on the session characteristic data;
determining a product to be recommended according to the session characteristic data and a preset product keyword mapping table, and pushing the product to be recommended and the label data;
and receiving product feedback data aiming at the product to be recommended and the label data, and adjusting a preset product keyword mapping table according to the product feedback data.
In one embodiment, the product keyword mapping table includes a score for each product keyword for each product;
determining a product to be recommended according to the session characteristic data and a preset product keyword mapping table, wherein the step of determining the product to be recommended comprises the following steps:
Extracting target product keywords in the session characteristic data, wherein the target product keywords are product keywords with word frequencies arranged in preset names;
searching a preset product keyword mapping table for a score of each product corresponding to a target product keyword;
constructing a target product keyword score vector according to the score of each product corresponding to the target product keyword;
and determining a product to be recommended according to the keyword score vector of the target product.
In one embodiment, adjusting the preset product keyword mapping table according to the product feedback data includes:
performing keyword recognition on the product feedback data, and determining interested products contained in the product feedback data;
searching the score of each target product keyword corresponding to the interested product from a preset product keyword mapping table;
and adjusting the score of the interesting product corresponding to each target product keyword according to a preset score updating rule.
In one embodiment, the method further comprises:
acquiring product browsing and purchasing behavior data of a session object;
determining a preference product of the session object according to the product browsing and purchasing behavior data;
searching a score of a preference product corresponding to each target product keyword from a preset product keyword mapping table;
And adjusting the score of the preference product corresponding to each target product keyword according to a preset score updating rule.
In one embodiment, extracting session feature data of the object session data comprises:
performing word segmentation processing on the object session data to obtain word segmentation results;
removing stop words in the word segmentation result;
and performing word frequency statistics on the word segmentation result after the stop words are removed to obtain session characteristic data.
In one embodiment, constructing tag data for the session object based on the session feature data comprises:
and calling a third-party word cloud picture generation tool, converting the session characteristic data into a word cloud picture according to preset word cloud picture style parameters, wherein the word cloud picture comprises the label data of the session object.
In a second aspect, the present application further provides a product data pushing apparatus, the apparatus including:
the data acquisition module is used for acquiring object session data;
the session data processing module is used for extracting session characteristic data of the object session data and constructing label data of the session object based on the session characteristic data;
the data pushing module is used for determining a product to be recommended according to the session characteristic data and a preset product keyword mapping table, and pushing the product to be recommended and the label data;
And the data feedback adjusting module is used for receiving product feedback data aiming at the product to be recommended and the label data and adjusting a preset product keyword mapping table according to the feedback data.
In one embodiment, the product keyword mapping table includes a score for each product keyword for each product;
the data pushing module is also used for extracting target product keywords in the session characteristic data, wherein the target product keywords are product keywords with word frequencies arranged in preset titles, searching the score of each product corresponding to the target product keywords from a preset product keyword mapping table, constructing a score vector of the target product keywords according to the scores of each product corresponding to the target product keywords, and determining a product to be recommended according to the score vector of the target product keywords.
In one embodiment, the data feedback adjustment module is further configured to perform keyword recognition on the feedback data, determine an interested product included in the feedback data, search a score of the interested product corresponding to each target product keyword from a preset product keyword mapping table, and adjust the score of the interested product corresponding to each target product keyword according to a preset score update rule.
In one embodiment, the data feedback adjustment module is further configured to obtain product browsing and purchasing behavior data of the session object, determine an interested product of the session object according to the product browsing and purchasing behavior data, search a score of the interested product corresponding to each target product keyword from a preset product keyword mapping table, and adjust a score of the interested product corresponding to each target product keyword according to a preset score update rule.
In one embodiment, the object session data processing module is further configured to perform word segmentation on the object session data to obtain word segmentation results, remove stop words in the word segmentation results, and perform word frequency statistics on the word segmentation results after the stop words are removed to obtain the session feature data.
In one embodiment, the object session data processing module is further configured to invoke a third-party word cloud graph generation tool, convert the session feature data into a word cloud graph according to preset word cloud graph style parameters, and the word cloud graph includes tag data of the object.
In a third aspect, the application also provides a computer device. The computer device comprises a memory storing a computer program and a processor implementing the following steps when executing the computer program:
Acquiring object session data;
extracting session characteristic data of the object session data, and constructing label data of the session object based on the session characteristic data;
determining a product to be recommended according to the session characteristic data and a preset product keyword mapping table, and pushing the product to be recommended and the label data;
and receiving product feedback data aiming at the product to be recommended and the label data, and adjusting a preset product keyword mapping table according to the product feedback data.
In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:
acquiring object session data;
extracting session characteristic data of the object session data, and constructing label data of the session object based on the session characteristic data;
determining a product to be recommended according to the session characteristic data and a preset product keyword mapping table, and pushing the product to be recommended and the label data;
and receiving product feedback data aiming at the product to be recommended and the label data, and adjusting a preset product keyword mapping table according to the product feedback data.
In a fifth aspect, the present application further provides a computer program product. The computer program product comprising a computer program which when executed by a processor performs the steps of:
Acquiring object session data;
extracting session characteristic data of the object session data, and constructing label data of the session object based on the session characteristic data;
determining a product to be recommended according to the session characteristic data and a preset product keyword mapping table, and pushing the product to be recommended and the label data;
and receiving product feedback data aiming at the product to be recommended and the label data, and adjusting a preset product keyword mapping table according to the product feedback data.
According to the product data pushing method, the device, the computer equipment, the storage medium and the computer program product, the conversation characteristic data of the conversation data of the object is extracted, the label data of the conversation object is constructed based on the conversation characteristic data, then, the product to be recommended is determined according to the conversation characteristic data and the preset product keyword mapping table, the product to be recommended and the label data are pushed, automatic recommendation of the product and automatic output of the label data are achieved, further, a service worker can further judge whether the pushed product and label data are accurate or not according to knowledge and professional service knowledge of the conversation object, product feedback data are given, product feedback data aiming at the product to be recommended and the label data are received, and the preset product keyword mapping table is adjusted according to the product feedback data. According to the scheme, the product to be recommended and the label data are simultaneously pushed based on the object session data and the preset product keyword mapping table, so that intelligent recommendation of the product and intelligent addition of the label data are realized, the complex operation that business personnel manually maintain object labels is omitted, and the error probability of manual label addition is reduced. Moreover, service personnel can judge whether the pushed product and label data are accurate or not by means of understanding of the session object and professional service knowledge, product feedback data are given, and a product keyword mapping table is adjusted according to the received product feedback data, so that the accuracy of product recommendation can be improved, and refined product recommendation is further realized.
Drawings
FIG. 1 is a diagram of an application environment of a product data push method in one embodiment;
FIG. 2 is a flowchart illustrating a method for pushing product data according to one embodiment;
FIG. 3 is a flowchart illustrating the steps of determining a product to be recommended in one embodiment;
FIG. 4 is a flowchart illustrating a step of adjusting a predetermined product keyword mapping table according to product feedback data in one embodiment;
FIG. 5 is a flowchart illustrating a step of adjusting a predetermined product keyword mapping table according to product feedback data in another embodiment;
FIG. 6 is a detailed flowchart of a product data pushing method according to another embodiment;
FIG. 7 is a block diagram showing a configuration of a product data push apparatus according to an embodiment;
FIG. 8 is a diagram illustrating an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
It should be noted that the user information, data and device information referred in the technical solution of the present application are information and data authorized by the user or fully authorized by each party, and the acquisition, storage, use, processing, etc. of the data all conform to the relevant regulations of the national laws and regulations.
The product data pushing method provided by the embodiment of the application can be applied to the application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The data storage system may store data that the server 104 needs to process. The data storage system may be integrated on the server 104, or may be located on the cloud or other network server. Specifically, the service person may send a product recommendation message to the server 104 through the terminal 102, the server 104 responds to the message to obtain object session data between the service person and the object, extract session feature data of the object session data, construct tag data of the session object based on the session feature data, determine a product to be recommended according to the session feature data and a preset product keyword mapping table, push the product to be recommended and the tag data, receive product feedback data for the product to be recommended and the tag data, and adjust the preset product keyword mapping table according to the product feedback data. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, internet of things devices and portable wearable devices, and the internet of things devices may be smart speakers, smart televisions, smart air conditioners, smart car-mounted devices, and the like. The portable wearable device can be a smart watch, a smart bracelet, a head-mounted device, and the like. The server 104 may be implemented as a stand-alone server or as a server cluster comprised of multiple servers.
In one embodiment, as shown in fig. 2, a product data pushing method is provided, which is described by taking the method as an example applied to the server 104 in fig. 1, and includes the following steps:
step 200, obtaining object session data.
In this embodiment, the object is described by taking a client as an example. The object conversation data refers to the conversation data of business personnel and clients. In this embodiment, the number of object sessions may be session data generated by the enterprise employee and the client communicating through the enterprise office management tool, and the session data may be understood as a chat record. When the method is specifically implemented, a service person can send a product recommendation message, the product recommendation message carries the identification data of a client and the identification data of the service person, and based on the identification data of the client and the identification data of the service person, a data pulling method provided by an office management tool official of an enterprise is called to pull the conversation data of the service person and the client. In practical application, the method can be used for pulling the session data of the enterprise employee and the client in the enterprise office management tool in batches and storing the session data in a persistent mode.
Step 400, extracting session characteristic data of the object session data, and constructing label data of the session object based on the session characteristic data.
After the object session data is obtained, because the session data usually includes product information and preference information of the object, in order to perform product recommendation in a targeted manner, the session feature data of the object session data may be extracted, and then, based on the session feature data, tag data of the session object is constructed. Specifically, the session object may refer to a customer, and the tag data includes basic information, personality characteristics, preference data, and possibly product of interest of the customer. In this embodiment, user portrait data of a client may be constructed based on session feature data.
And step 600, determining a product to be recommended according to the conversation characteristic data and a preset product keyword mapping table, and pushing the product to be recommended and the label data.
The product keyword mapping table contains mapping relations of products, label data and product keywords in the product library. Specifically, the product keywords may correspond to specific products, each product type corresponds to a plurality of product keywords, and one product keyword may correspond to a plurality of product types. The extracted conversation feature data comprises a plurality of product keywords, which can be the product keywords extracted from the conversation feature data, then the product keywords are searched in a product keyword mapping table according to the product keywords, and then the product to be recommended is determined according to the searched information. Further, products to be recommended and tag data are pushed, so that enterprise staff can recommend the products to clients.
Step 800, receiving product feedback data aiming at the product to be recommended and the label data, and adjusting a preset product keyword mapping table according to the product feedback data.
After the product to be recommended and the tag data are pushed, the business personnel can confirm the accuracy of the pushed product by combining the tag data, namely, by combining the knowledge of the business personnel on the client, the tag data and the professional knowledge, determine whether the pushed product meets the client requirements, and if the pushed product is the product in which the client is interested, the product feedback data can be as follows: "product a is an interesting product", if the pushed product does not meet the customer's requirement, the recommendation accuracy is not high, and the correct product that the customer may be interested in is input, and the product feedback data may be "product a is not an interesting product, and the interesting product may be product B". After product feedback data fed back by service personnel is received, a preset product keyword mapping table can be adjusted according to the product feedback data, and data in the product keyword mapping table is updated, so that the product recommendation accuracy is improved. Moreover, iterative loop can be performed according to the above manner, and after the product to be recommended is pushed every time, business personnel all feed back corresponding product feedback data so as to adjust the product keyword mapping table in time.
According to the product data pushing method, conversation characteristic data of object conversation data are extracted, label data of a conversation object are constructed based on the conversation characteristic data, then products to be recommended are determined according to the conversation characteristic data and a preset product keyword mapping table, the products to be recommended and the label data are pushed, automatic recommendation of the products and automatic output of the label data are achieved, further, service personnel can further judge whether the pushed products and the label data are accurate or not according to knowledge and professional service knowledge of the conversation object, product feedback data are given, product feedback data aiming at the products to be recommended and the label data are received, and the preset product keyword mapping table is adjusted according to the product feedback data. According to the scheme, the product to be recommended and the label data are simultaneously pushed based on the object session data and the preset product keyword mapping table, intelligent recommendation of the product and intelligent addition of the label data are achieved, tedious operations of service personnel for maintaining the object label manually are omitted, and the probability of errors in manual label addition is reduced. Moreover, service personnel can judge whether the pushed product and label data are accurate or not by means of the knowledge of the session object and professional service knowledge, provide product feedback data, and adjust a product keyword mapping table according to the received product feedback data, so that the accuracy of product recommendation can be improved, and refined product recommendation can be realized.
As shown in FIG. 3, in one embodiment, the product keyword mapping table includes a score for each product keyword for each product;
step 600 comprises:
and step 620, extracting target product keywords in the session characteristic data, wherein the target product keywords are product keywords with word frequencies arranged in preset names.
In step 640, the score of each product corresponding to the target product keyword is searched in a preset product keyword mapping table.
And 660, constructing a target product keyword score vector according to the score of each product corresponding to the target product keyword.
And step 680, determining the product to be recommended according to the keyword score vector of the target product.
In practical application, at the beginning of construction of the product keyword mapping table, a service person sets a plurality of corresponding product keywords for each product, and each product keyword has a specific score for each product, for example, 100 products exist in a product library, a product keyword a has a corresponding score for each product in the 100 products, that is, the product keyword score vector of the product keyword a is a 100-dimensional vector, and the score is in direct proportion to the relevancy of the product keyword. If the product keyword a is not directly linked to the product XX, the score of the product keyword a relative to the product XX may be initialized to 0, and if the correlation between the product keyword a and the product XX is high, the score of the product keyword a relative to the product XX may be initialized to 5. In this embodiment, because the session feature data includes a plurality of product keywords, product keywords whose frequency ranks in the top 5 may be extracted from the session feature data, the extracted 5 product keywords are used as target product keywords, then scores of the target product keywords corresponding to each product are searched in a preset product keyword mapping table, a target product keyword score vector is constructed according to the scores of the target product keywords corresponding to each product, then, the sum of the scores of all the target product keywords for each product is calculated according to the target product keyword score vector, a total score of each product in units of the target product keywords is obtained, and then, the product whose score ranks in the top five is determined as the product to be recommended. For example, if there are n products in the product library, the score for each word for n products can be expressed as a product keyword score vector VC: VC = [ C1, C2, \8230; cn-1, cn ], wherein Cn represents the score of the nth product corresponding to the word C, and the total score of each product is calculated by the five extracted words C1, C2, C3, C4 and C5 with the highest word frequency. The total score Scoren for the nth product is: scoren = C1n + C2n + C3n + C4n + C5n, the first five products with the highest scores will be determined as the products to be recommended. In the embodiment, the target product keywords are screened out according to the word frequency, and then the product to be recommended is determined according to the value of each product corresponding to the target product keywords in the product keyword mapping table, so that the screened product to be recommended can better meet the requirements of customers, and the accuracy is higher.
As shown in fig. 4, in an embodiment, adjusting the preset product keyword mapping table according to the product feedback data includes:
and step 820, performing keyword recognition on the product feedback data, and determining the interested product contained in the product feedback data.
Step 840, finding the score of the interested product corresponding to each target product keyword in a preset product keyword mapping table.
And step 860, adjusting the score of each target product keyword corresponding to the interested product according to a preset score updating rule.
The acquisition mode of the product feedback data can be manually input by service personnel, or can be determined by a background according to the product browsing and purchasing behavior data of a client. In this embodiment, the product feedback data is manually input by a service person as an example. It can be known from the above embodiment that each target product keyword has a score for each product, and after product feedback data fed back by service personnel is received, keyword recognition may be performed on the product feedback data to determine an interested product included in the product feedback data, then, a score of the interested product corresponding to each target product keyword is searched in a preset product keyword mapping table, and then, the score of the interested product corresponding to each target product keyword is updated according to an update coefficient in a score update rule. For example, in the process of interacting with the customer, the service personnel deduces the product in which the customer is interested, and after confirming that the pushed product is not the product in which the customer is interested after receiving the pushed product and the tag data, the service personnel inputs the deduced product in which the customer is interested and receives the product in which the service personnel is interested After the product feedback data fed back by the service personnel is received, keyword recognition can be carried out on the product feedback data to determine the interesting product input by the service personnel, and then according to a preset score updating rule, the scores of the keywords of the first five target products with the highest word frequency in the chat records of the client on the interesting product are updated. If the interested product input by the service personnel is the mth product in the product library, five words C with the highest word frequency in the chat records of the service personnel and the client 1 、C 2 、C 3 、C 4 、C 5 Score for mth product C1 m 、C2 m 、C3 m 、C4 m 、C5 m Updating the rules according to a predetermined score, C1 m 、C2 m 、C3 m 、C4 m 、C5 m Are respectively updated to alpha C1 m 、αC2 m 、αC3 m 、αC4 m 、αC5 m And alpha is an adjusting factor which is greater than or equal to 1, so that the scores of the five target product keywords for the mth product are improved, and the relevance of the product keywords and the product is strengthened. In the embodiment, the score of the interesting product corresponding to each target product keyword is updated according to the preset score updating rule, the product keyword mapping table can be simply and quickly updated, and the accuracy of subsequent product recommendation can be improved.
As shown in fig. 5, in one embodiment, the method further comprises:
step 900, acquiring product browsing and purchasing behavior data of the session object.
Step 920, determining a preferred product of the session object according to the product browsing and purchasing behavior data.
In step 940, the score of the preferred product corresponding to each target product keyword is searched in a preset product keyword mapping table.
And step 960, adjusting the score of the preference product corresponding to each target product keyword according to a preset score updating rule.
The product browsing and purchasing behavior data includes product browsing data and product purchasing data. Specifically, the platform may be a platform such as a mall system and a product system of an enterprise, which is buried in the platformDetermining a preferred product purchased or browsed for many times by a client after detecting that the client browses a certain product for many times or purchases a certain product according to the product browsing and purchasing behavior data, then searching a score of the preferred product corresponding to each target product keyword in a preset product keyword mapping table, similarly, adjusting scores of the preferred products corresponding to the first five target product keywords with the highest word frequency in the chat records of the client according to a preset score updating rule, and if the preferred product determined according to the product browsing and purchasing behavior of the client is the nth product in the product library, adjusting five words C with the highest word frequency in the chat records of the service personnel and the client according to the preset score updating rule 1 、C 2 、C 3 、C 4 、C 5 Score C1 for mth product n 、C2 n 、C3 n 、C4 n 、C5 n Updating the rules according to a predetermined score, C1 n 、C2 n 、C3 n 、C4 n 、C5 n Are respectively updated to alpha C1 n 、αC2 n 、αC3 n 、αC4 n 、αC5 n Wherein alpha is an adjustment factor of more than or equal to 1. In the embodiment, the preference product of the client is automatically determined according to the product browsing and purchasing behavior data of the client, the score of the preference product corresponding to each target product keyword is updated according to the preset score updating rule, the product keyword mapping table can be simply and quickly updated, the accuracy of subsequent product recommendation can be improved, and the product which is more in line with the preference of the client can be recommended.
In one embodiment, extracting session feature data of the object session data comprises: and performing word segmentation processing on the object conversation data to obtain word segmentation results, eliminating stop words in the word segmentation results, and performing word frequency statistics on the word segmentation results after the stop words are eliminated to obtain conversation characteristic data.
In this embodiment, a Jieba word segmentation method may be adopted to perform word segmentation on the object session data to obtain word segmentation results, and then, according to a preset stop word dictionary, stop words in the word segmentation results are removed, and meaningless words, such as "ones", "bars", "woollens", and the like, are filtered, and simultaneously, emoji expressions, punctuation marks, and other information are filtered. Then, performing word frequency statistics on the word segmentation result after the stop words are removed, calculating a word frequency-reverse file frequency tf-idfi value of the word segmentation result after data cleaning, then performing descending sequencing on word groups according to the tf-idfi value to obtain a word frequency statistical list, and taking the word frequency statistical list as session characteristic data. It is understood that, in other embodiments, other word segmentation methods such as a word segmentation method based on string matching may also be used, which may be determined according to actual situations and is not limited herein. In the embodiment, word segmentation processing and word frequency statistics are performed on the object session data, and the word frequency statistical result data is used as the session characteristic data, so that a data basis can be provided for constructing the tag data of the session object, and the tag data can be added conveniently.
In one embodiment, constructing the tag data for the session object comprises: and calling a third-party word cloud picture generation tool, and converting the session characteristic data into a word cloud picture according to preset word cloud picture style parameters, wherein the word cloud picture comprises the label data of the session object.
The word cloud picture, also called character cloud, visually highlights the 'key word' with high frequency of occurrence in the text, and the more the occurrence frequency is, the larger the displayed font is, and the more prominent the key word is, the more important the representation is. In this embodiment, the tag data of the session object is embodied in the form of a word cloud. Specifically, a third-party word cloud picture generation tool such as a WordCloud word cloud library is called, and then the session feature data is converted into a word cloud picture according to preset word cloud picture style parameters, wherein feature keywords of the session object are highlighted in the word cloud picture. Specifically, the generation process of the word cloud graph may include: determining the shape, line color and filling color of the base map according to preset cloud picture style parameters, determining the size of a canvas according to the shape of the base map, and drawing the base map on the canvas according to the line color and filling color of the base map so as to generate the base map with a binary format. After a binary base map is generated, determining word areas capable of being allocated to keywords in the area of the base map according to positive correlation relations between word frequencies of the keywords and word areas of the keywords in the session characteristic data, determining font types and font colors of the keywords, converting the keywords into font maps which accord with the font types and the font colors and occupy the word areas, and merging the font maps into the base map to generate a word cloud map. In the embodiment, the tag data of the session object is constructed in a word cloud picture generating mode, so that a large amount of low-frequency and low-quality text information can be filtered, business personnel can intuitively draw the characteristics of the session object, and products meeting the requirements of the session object can be quickly deduced.
For a clearer explanation of the product data pushing method provided by the present application, the following description is made with reference to fig. 6 and a specific embodiment, where the specific embodiment may include the following steps:
step 1: and calling a data pulling method provided by an office management tool official of the enterprise to pull the session data of the business personnel and the client.
Step 2: performing word segmentation processing on the object session data to obtain word segmentation results, eliminating stop words in the word segmentation results, and performing word frequency statistics on the word segmentation results after the stop words are eliminated to obtain session characteristic data.
And step 3: and calling a WordCloud word cloud library, and converting the session characteristic data into a word cloud picture according to a preset word cloud picture style parameter, wherein the word cloud picture comprises the label data of the client.
And 4, step 4: extracting target product keywords in the session characteristic data, wherein the target product keywords are product keywords with word frequencies arranged in a preset ranking, searching a score of each product corresponding to the target product keywords in a preset product keyword mapping table, constructing a target product keyword score vector according to the scores of each product corresponding to the target product keywords, screening the top 5 products with the highest score according to the target product keyword score vector, determining the products to be recommended, and pushing the products to be recommended and a word cloud picture. For example, where there are n products in the product library, the score for each word for n products may be expressed as a product keyword score vector VC: VC = [ C1, C2, \8230; cn-1, cn ], wherein Cn represents the score of the nth product corresponding to the word C, and the total score of each product is calculated by the five extracted words C1, C2, C3, C4 and C5 with the highest word frequency. The total score Scoren for the nth product is: scoren = C1n + C2n + C3n + C4n + C5n, the first five products with the highest scores will be determined as the products to be recommended
And 5: receiving product feedback data aiming at the product to be recommended and the label data, performing keyword identification on the product feedback data, determining interested products contained in the product feedback data, searching the score of the interested product corresponding to each target product keyword in a preset product keyword mapping table, and adjusting the score of the interested product corresponding to each target product keyword according to a preset score updating rule. For example, the product of interest entered by the business person is the mth product in the product library, and the five words C with the highest word frequency in the chat records of the business person and the client 1 、C 2 、C 3 、C 4 、C 5 Score of C1 for mth product m 、C2 m 、C3 m 、C4 m 、C5 m Updating the rules according to a predetermined score, C1 m 、C2 m 、C3 m 、C4 m 、C5 m Are respectively updated to alpha C1 m 、αC2 m 、αC3 m 、αC4 m 、αC5 m And the alpha is an adjusting factor which is greater than or equal to 1, so that the scores of the five target product keywords for the mth product are improved, and the relevance of the product keywords and the product is strengthened.
Step 6: and acquiring product browsing and purchasing behavior data of the session object, and determining a preference product of the session object according to the product browsing and purchasing behavior data. And searching the score of the preference product corresponding to each target product keyword in a preset product keyword mapping table. And adjusting the score of the preference product corresponding to each target product keyword according to a preset score updating rule.
It should be understood that, although the steps in the flowcharts related to the embodiments as described above are sequentially displayed as indicated by arrows, the steps are not necessarily performed sequentially as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a part of the steps in the flowcharts related to the embodiments described above may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the execution order of the steps or stages is not necessarily sequential, but may be rotated or alternated with other steps or at least a part of the steps or stages in other steps.
Based on the same inventive concept, the embodiment of the present application further provides a product data pushing apparatus for implementing the above-mentioned product data pushing method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so specific limitations in one or more embodiments of the product data pushing device provided below can be referred to the limitations in the above product data pushing method, and are not described herein again.
In one embodiment, as shown in fig. 7, there is provided a product data pushing apparatus including: a data acquisition module 710, a session data processing module 720, a data push module 730, and a data feedback adjustment module 740, wherein:
a data obtaining module 710, configured to obtain object session data;
the session data processing module 720 is used for extracting session characteristic data of the object session data and constructing tag data of the session object based on the session characteristic data;
the data pushing module 730 is used for determining a product to be recommended according to the session characteristic data and a preset product keyword mapping table, and pushing the product to be recommended and the label data;
and the data feedback adjusting module 740 is configured to receive product feedback data for the product to be recommended and the label data, and adjust the preset product keyword mapping table according to the feedback data.
In the product data pushing device, the conversation characteristic data of the conversation data of the object is extracted, the label data of the conversation object is constructed based on the conversation characteristic data, then, the product to be recommended is determined according to the conversation characteristic data and the preset product keyword mapping table, the product to be recommended and the label data are pushed, automatic recommendation of the product and automatic output of the label data are achieved, further, a service worker can further judge whether the pushed product and label data are accurate or not according to the knowledge and professional service knowledge of the conversation object, product feedback data are given, product feedback data aiming at the product to be recommended and the label data are received, and the preset product keyword mapping table is adjusted according to the product feedback data. According to the scheme, the product to be recommended and the label data are simultaneously pushed based on the object session data and the preset product keyword mapping table, so that intelligent recommendation of the product and intelligent addition of the label data are realized, the complex operation that business personnel manually maintain object labels is omitted, and the error probability of manual label addition is reduced. Moreover, service personnel can judge whether the pushed product and label data are accurate or not by means of the knowledge of the session object and professional service knowledge, further give product feedback data, and adjust a product keyword mapping table according to the received product feedback data, so that the accuracy of product recommendation can be improved, and refined product recommendation is realized.
In one embodiment, the product keyword mapping table includes a score for each product keyword for each product;
the data pushing module 730 is further configured to extract target product keywords in the session feature data, where the target product keywords are product keywords with word frequencies arranged in preset ranks, search a preset product keyword mapping table for scores of each product corresponding to the target product keywords, construct a target product keyword score vector according to the scores of each product corresponding to the target product keywords, and determine a product to be recommended according to the target product keyword score vector.
In an embodiment, the data feedback adjustment module 740 is further configured to perform keyword recognition on the feedback data, determine the interested products included in the feedback data, search the score of the interested product corresponding to each target product keyword from a preset product keyword mapping table, and adjust the score of the interested product corresponding to each target product keyword according to a preset score update rule.
In one embodiment, the data feedback adjusting module 740 is further configured to obtain product browsing and purchasing behavior data of the session object, determine an interested product of the session object according to the product browsing and purchasing behavior data, find a score of the interested product corresponding to each target product keyword from a preset product keyword mapping table, and adjust the score of the interested product corresponding to each target product keyword according to a preset score updating rule.
In an embodiment, the object session data processing module 720 is further configured to perform word segmentation on the object session data to obtain a word segmentation result, remove stop words in the word segmentation result, and perform word frequency statistics on the word segmentation result after removing the stop words to obtain session feature data.
In one embodiment, the object session data processing module 720 is further configured to invoke a third party word cloud graph generation tool to convert the session feature data into a word cloud graph according to a preset word cloud graph style parameter, where the word cloud graph includes tag data of the object.
The modules in the product data pushing device can be wholly or partially implemented by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 8. The computer device includes a processor, a memory, an Input/Output interface (I/O for short), and a communication interface. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface is connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing data such as object session data, label data, product feedback data and the like. The input/output interface of the computer device is used for exchanging information between the processor and an external device. The communication interface of the computer device is used for connecting and communicating with an external terminal through a network. The computer program is executed by a processor to implement a product data pushing method.
Those skilled in the art will appreciate that the architecture shown in fig. 8 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided, which includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the product data pushing method when executing the computer program.
In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and the computer program, when executed by a processor, implements the steps in the above product data pushing method.
In one embodiment, a computer program product is provided, comprising a computer program that when executed by a processor implements the steps in the above-described product data pushing method.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include a Read-Only Memory (ROM), a magnetic tape, a floppy disk, a flash Memory, an optical Memory, a high-density embedded nonvolatile Memory, a resistive Random Access Memory (ReRAM), a Magnetic Random Access Memory (MRAM), a Ferroelectric Random Access Memory (FRAM), a Phase Change Memory (PCM), a graphene Memory, and the like. Volatile Memory can include Random Access Memory (RAM), external cache Memory, and the like. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others. The databases referred to in various embodiments provided herein may include at least one of relational and non-relational databases. The non-relational database may include, but is not limited to, a block chain based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, quantum computing based data processing logic devices, etc., without limitation.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.
Claims (15)
1. A method of pushing product data, the method comprising:
acquiring object session data;
extracting session characteristic data of the object session data, and constructing tag data of a session object based on the session characteristic data;
determining a product to be recommended according to the conversation feature data and a preset product keyword mapping table, and pushing the product to be recommended and the tag data;
And receiving product feedback data aiming at the product to be recommended and the label data, and adjusting the preset product keyword mapping table according to the product feedback data.
2. The method of claim 1, wherein the product keyword mapping table comprises a score for each product keyword for each product;
the step of determining the product to be recommended according to the session feature data and a preset product keyword mapping table comprises the following steps:
extracting target product keywords in the session characteristic data, wherein the target product keywords are product keywords with word frequencies arranged in a preset ranking;
searching a score of each product corresponding to the target product keyword in the preset product keyword mapping table;
constructing a target product keyword score vector according to the score of each product corresponding to the target product keyword;
and determining a product to be recommended according to the keyword score vector of the target product.
3. The product data pushing method of claim 2, wherein adjusting the preset product keyword mapping table according to product feedback data comprises:
performing keyword recognition on the product feedback data, and determining interested products contained in the product feedback data;
Searching a score of each target product keyword corresponding to the interesting product from the preset product keyword mapping table;
and adjusting the score of each target product keyword corresponding to the interested product according to a preset score updating rule.
4. The product data pushing method of claim 2, further comprising:
acquiring product browsing and purchasing behavior data of the session object;
determining a preference product of the session object according to the product browsing and purchasing behavior data;
searching a score of each target product keyword corresponding to the preference product from the preset product keyword mapping table;
and adjusting the score of each target product keyword corresponding to the preference product according to a preset score updating rule.
5. The product data pushing method according to claim 1, wherein the extracting session feature data of the object session data includes:
performing word segmentation processing on the object session data to obtain word segmentation results;
removing stop words in the word segmentation result;
and performing word frequency statistics on the word segmentation result after the stop words are removed to obtain conversation characteristic data.
6. The product data pushing method according to claim 1, wherein the constructing tag data of a session object based on the session feature data comprises:
and calling a third-party word cloud picture generation tool, and converting the session characteristic data into a word cloud picture according to preset word cloud picture style parameters, wherein the word cloud picture comprises the label data of the session object.
7. A product data push apparatus, the apparatus comprising:
the data acquisition module is used for acquiring object session data;
the session data processing module is used for extracting session characteristic data of the object session data and constructing label data of a session object based on the session characteristic data;
the data pushing module is used for determining a product to be recommended according to the session characteristic data and a preset product keyword mapping table, and pushing the product to be recommended and the label data;
and the data feedback adjusting module is used for receiving product feedback data aiming at the product to be recommended and the label data and adjusting the preset product keyword mapping table according to the feedback data.
8. The apparatus of claim 7, wherein the product keyword mapping table comprises a score for each product keyword;
The data pushing module is further used for extracting target product keywords in the session characteristic data, the target product keywords are product keywords with word frequencies arranged in preset names, scores of products corresponding to the target product keywords are searched from a preset product keyword mapping table, score vectors of the target product keywords are constructed according to the scores of the products corresponding to the target product keywords, and products to be recommended are determined according to the score vectors of the target product keywords.
9. The device for pushing product data according to claim 8, wherein the data feedback adjustment module is further configured to perform keyword recognition on the feedback data, determine an interested product included in the feedback data, search a score of each target product keyword corresponding to the interested product from the preset product keyword mapping table, and adjust the score of each target product keyword corresponding to the interested product according to a preset score update rule.
10. The product data pushing device according to claim 8, wherein the data feedback adjusting module is further configured to obtain product browsing and purchasing behavior data of the session object, determine an interested product of the session object according to the product browsing and purchasing behavior data, find a score of each target product keyword corresponding to the interested product from the preset product keyword mapping table, and adjust the score of each target product keyword corresponding to the interested product according to a preset score updating rule.
11. The product data pushing device according to claim 7, wherein the object session data processing module is further configured to perform word segmentation on the object session data to obtain word segmentation results, remove stop words in the word segmentation results, and perform word frequency statistics on the word segmentation results after the stop words are removed to obtain session feature data.
12. The product data pushing device according to claim 7, wherein the object session data processing module is further configured to invoke a third-party word cloud graph generation tool, and convert the session feature data into a word cloud graph according to preset word cloud graph style parameters, where the word cloud graph includes tag data of the object.
13. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 6.
14. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 6.
15. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211173757.1A CN115393024A (en) | 2022-09-26 | 2022-09-26 | Product data pushing method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211173757.1A CN115393024A (en) | 2022-09-26 | 2022-09-26 | Product data pushing method and device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115393024A true CN115393024A (en) | 2022-11-25 |
Family
ID=84128161
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211173757.1A Pending CN115393024A (en) | 2022-09-26 | 2022-09-26 | Product data pushing method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115393024A (en) |
-
2022
- 2022-09-26 CN CN202211173757.1A patent/CN115393024A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111797210A (en) | Information recommendation method, device and equipment based on user portrait and storage medium | |
| US20140172566A1 (en) | Matching of advertising sources and keyword sets in online commerce platforms | |
| US10248712B1 (en) | Generating a set of representative items using a maximum-set-coverage selection strategy | |
| US20170286970A1 (en) | Answer-suggestion system for automatically resolving customer requests | |
| US20210133203A1 (en) | System and method for converting user data from disparate sources to bitmap data | |
| CN114065750A (en) | Commodity information matching and publishing method and device, equipment, medium and product thereof | |
| CN107609192A (en) | The supplement searching method and device of a kind of search engine | |
| CN112907334A (en) | Object recommendation method and device | |
| CN111428100A (en) | Data retrieval method and device, electronic equipment and computer-readable storage medium | |
| CN116823404A (en) | Commodity combination recommendation method, device, equipment and medium thereof | |
| US10114885B1 (en) | Generating a set of representative items using a clustering-selection strategy | |
| CN116597443A (en) | Material tag processing method and device, electronic equipment and medium | |
| CN115936805A (en) | Commodity recommendation method, commodity recommendation device, commodity recommendation equipment and commodity recommendation medium | |
| CN113343684B (en) | Core product word recognition method, device, computer equipment and storage medium | |
| CN115393024A (en) | Product data pushing method and device, computer equipment and storage medium | |
| CN114971766A (en) | Commodity recommendation method and device, equipment, medium and product thereof | |
| JP2015088089A (en) | Recommendation server and recommendation method | |
| CN113762992B (en) | Method and device for processing data | |
| CN113327145B (en) | Article recommendation method and device | |
| CN113780827A (en) | Article screening method and device, electronic equipment and computer readable medium | |
| CN109978645B (en) | Data recommendation method and device | |
| EP3023887A1 (en) | Apparatus, method and medium for recommending contextual information | |
| KR101985603B1 (en) | Recommendation method based on tripartite graph | |
| CN111339124A (en) | Data display method and device, electronic equipment and computer readable medium | |
| CN113821722B (en) | Data processing method, recommending device, electronic equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |