CN109727092A - Products Show method, apparatus, computer equipment and storage medium based on AI - Google Patents
Products Show method, apparatus, computer equipment and storage medium based on AI Download PDFInfo
- Publication number
- CN109727092A CN109727092A CN201811537638.3A CN201811537638A CN109727092A CN 109727092 A CN109727092 A CN 109727092A CN 201811537638 A CN201811537638 A CN 201811537638A CN 109727092 A CN109727092 A CN 109727092A
- Authority
- CN
- China
- Prior art keywords
- product
- target
- recommended
- keyword
- client
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention discloses a kind of Products Show method, apparatus, computer equipment and storage medium based on AI, and this method includes the target answer voice data for obtaining client and being sent based on asked questions；Target answer voice data is analyzed and processed using Natural Language Processing Models, obtains product purchasing demand corresponding with target answer voice data；Product purchasing demand is matched with the attribute keywords in multidimensional data table, obtains at least one corresponding product to be recommended of attribute keywords of successful match；Calculate each product to be recommended matching degree corresponding with product purchasing demand；Based on matching degree, descending sort is carried out at least one product to be recommended, the product to be recommended of preset quantity is determined as target product；Database is searched based on target product, obtains analysis data corresponding with target product, by target product and corresponding analysis data feedback to client, the important information of recommended products cannot embody when solving the problem of to recommend product.
Description
Technical field
The present invention relates to the field AI more particularly to a kind of Products Show method, apparatus based on AI, computer equipment and deposit
Storage media.
Background technique
In existing product recommended method, Products Show is mainly carried out by the self-characteristic of client.For example, passing through client
Age and gender etc. carry out Products Show, or Products Show is carried out by the browsing record of client, so that recommended products is not
Meet client's purchasing demand.When recommended products is more or the clause of each recommended products is more, client can not determine each push away
It recommends the important information of product, or about important informations such as similarities and differences between at least two recommended products, also needs client voluntarily
Analysis and understanding is carried out to the important information of recommended products, leads to cannot embodying for the important information of recommended products, so that its is right
The attraction of user is inadequate, and user can not be attracted to use or buy recommended products.
Summary of the invention
The embodiment of the present invention provides a kind of Products Show method, apparatus, computer equipment and storage medium based on AI, with
Solve the problems, such as cannot embodying for the important information of current recommended products.
A kind of Products Show method based on AI, comprising:
Obtain the target answer voice data that client is sent based on asked questions；
The target answer voice data is analyzed and processed using Natural Language Processing Models, is obtained and the target
The corresponding product purchasing demand of answer voice data；
The product purchasing demand is matched with the attribute keywords in multidimensional data table, obtains the category of successful match
Property at least one corresponding product to be recommended of keyword；
Calculate each product to be recommended matching degree corresponding with the product purchasing demand；
Based on the matching degree, descending sort is carried out at least one described product to be recommended, it will be described in preset quantity
Product to be recommended is determined as target product；
Database is searched based on each target product, obtains analysis data corresponding with each target product,
Give each target product and the corresponding analysis data feedback to the client.
A kind of Products Show device based on AI, comprising:
Answer obtains module, the target answer voice data sent for obtaining client based on asked questions；
Requirement extract module, for being carried out at analysis using Natural Language Processing Models to the target answer voice data
Reason obtains product purchasing demand corresponding with the target answer voice data；
Product to be recommended obtains module, for by the attribute keywords in the product purchasing demand and multidimensional data table into
Row matching, obtains at least one corresponding product to be recommended of attribute keywords of successful match；
Matching degree computing module, for calculating the matching corresponding with the product purchasing demand of each product to be recommended
Degree；
Target product obtains module, for being based on the matching degree, carries out descending at least one described product to be recommended
Sequence, is determined as target product for the product to be recommended of preset quantity；
Data feedback module is obtained and is produced with each target for searching database based on each target product
The corresponding analysis data of product give each target product and the corresponding analysis data feedback to the client.
A kind of computer equipment, including memory, processor and storage are in the memory and can be in the processing
The computer program run on device, the processor realize the above-mentioned Products Show side based on AI when executing the computer program
The step of method.
A kind of computer readable storage medium, the computer-readable recording medium storage have computer program, the meter
Calculation machine program realizes the step of above-mentioned Products Show method based on AI when being executed by processor.
It is above-mentioned that a kind of Products Show method, apparatus, computer equipment and storage medium based on AI is provided, using nature language
The target answer voice data corresponding with asked questions that speech processing model sends client is analyzed and processed, and is realized to mesh
It marks answer voice data and carries out intellectual analysis, with quick obtaining to product purchasing demand, without being manually entered answer.Product is purchased
Demand is bought to be matched with the attribute keywords in multidimensional data table, with realize obtain largely meet product purchasing demand wait push away
Recommend product.By calculating each product to be recommended matching degree corresponding with product purchasing demand, by the matching degree of preset quantity compared with
High product to be recommended is determined as target product, to improve the accuracy of target product recommendation.It obtains corresponding with target product
Data are analyzed, by each target product and corresponding analysis data feedback to client, client can be comprehensive according to analysis data
Solve target product so that the important information for the target product that recommended products is recommended is embodied, with improve recommended products to
The attraction at family.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below by institute in the description to the embodiment of the present invention
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention
Example, for those of ordinary skill in the art, without any creative labor, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is the application environment schematic diagram of the Products Show method in one embodiment of the invention based on AI；
Fig. 2 is the flow chart of the Products Show method in one embodiment of the invention based on AI；
Fig. 3 is the flow chart of the Products Show method in one embodiment of the invention based on AI；
Fig. 4 is the flow chart of the Products Show method in one embodiment of the invention based on AI；
Fig. 5 is the flow chart of the Products Show method in one embodiment of the invention based on AI；
Fig. 6 is the flow chart of the Products Show method in one embodiment of the invention based on AI；
Fig. 7 is the flow chart of the Products Show method in one embodiment of the invention based on AI；
Fig. 8 is the functional block diagram of the Products Show device in one embodiment of the invention based on AI；
Fig. 9 is a schematic diagram of computer equipment in one embodiment of the invention.
Specific embodiment
Below by the attached drawing in knot and the embodiment of the present invention, technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
Products Show method provided in an embodiment of the present invention based on AI, can be applicable in the application environment such as Fig. 1, the base
It is applied in Products Show system in the Products Show method of AI, which includes client and server-side, wherein
Client is communicated by network with server-side.The Products Show method based on AI is particularly applicable in Products Show APP's
In server-side, server-side sends asked questions to client, and client is based on asked questions to server-side feedback target answer language
Sound data handle target answer voice data by Natural Language Processing Models, obtain from target answer voice data
Product purchasing demand is taken, and corresponding target product is got according to product purchasing demand, and obtain corresponding with target product
Data are analyzed, the analysis data and target product are fed back into client, so that client becomes more apparent upon target by analyzing data
Product, so that the important information for the target product recommended is embodied.Wherein, client can be, but not limited to various personal meters
Calculation machine, laptop, smart phone, tablet computer and portable wearable device.Server-side can use independent server-side
The either server-side cluster of multiple server-sides composition is realized.
In one embodiment, it as shown in Fig. 2, providing a kind of Products Show method based on AI, is applied in this way in Fig. 1
In server-side for be illustrated, specifically comprise the following steps:
S10: the target answer voice data that client is sent based on asked questions is obtained.
Wherein, target answer voice data refer to client by voice mode sent to server-side with asked questions pair
The answer answered.
Specifically, client can carry out man-machine interactively with server-side, and server-side is from preset asked questions library to client
End sends asked questions.Server-side, which is provided, obtains interface with voice, and client is based on the voice and obtains interface, anti-to server-side
Present target answer voice data corresponding with asked questions.It is to be appreciated that should include client couple in target answer voice data
The product purchasing demand answered.Asked questions are contained at least two in preset asked questions library.Wherein, asked questions, which can be, draws
The asked questions of conduction, so that client sends the target answer voice data comprising product purchasing demand based on the asked questions.
Server-side gets the target answer voice data of client transmission based on asked questions, so as to subsequent according to target answer voice
Data acquisition is manually entered target answer to the corresponding product purchasing demand of client, and without client, to increase the groups such as old age
The purchasing demand of body.
S20: being analyzed and processed target answer voice data using Natural Language Processing Models, obtains and target answer
The corresponding product purchasing demand of voice data.
Wherein, product purchasing demand refers to the purchasing demand information extracted from target answer voice data, specifically may be used
To be the demand keyword extracted from target answer voice data.Wherein, demand keyword refers to the mesh sent from client
The keyword extracted in mark answer voice data, constitutes product purchasing demand by one or more demand keywords.Natural language
Handling (natural language processing, abbreviation NLP) model is to be able to achieve between people and computer to use natural language
Carry out the various theory and methods of efficient communication.Natural Language Processing Models can be used for target answer voice data carrying out voice
It is converted into the processing model of writing text, including two processes of natural language understanding and spatial term.For example, can pass through
Hidden Markov model is converted into vocabulary (natural language understanding process) user voice data of acquisition, then passes through morphology point
The technologies such as analysis, syntactic analysis, semantic analysis and text generation generate user answer text (spatial term process), then to
Text Feature Extraction key message is answered at family, to get product purchasing demand.Wherein, key message can be understood as demand key
Word.Optionally, Natural Language Processing Models can use the training pattern of deep learning, pass through keyword in a large amount of voice data
It is annotated, forms speech samples, be input in training pattern and be trained, obtain training pattern according to voice data
Corresponding text data, to get Natural Language Processing Models.
Specifically, by the way that target answer voice data to be input in Natural Language Processing Models, at natural language
It manages model and semantic analysis and understanding is carried out to target answer voice data, to get the corresponding product purchasing demand of client.
Wherein, product purchasing demand can be age, gender, the requirements of support and insurance amount etc..Wherein, the requirements of support can be " health
Insurance ", " casualty insurance ", " life insurance " and " juvenile's insurance " etc..
Further, in the server-side of Products Show APP, also there is guiding function module, in the guiding function module
Sentence pattern example is provided, for example, " I wants xx product "；For another example the general sentence such as " I will buy xx insurance, and it is low to insure amount "
The sentence pattern example is shown the display page in Products Show APP, or by man-machine interaction mode, is broadcast by client by type
Sentence pattern example is put, guidance client is acquired by client and sends target answer voice data, target answer language according to sentence pattern example
In sound data include client's purchase intention, by Natural Language Processing Models to target answer voice data carry out semantic analysis and
Understand, gets user and answer text, then answered from user and obtain demand keyword in text, that is, get and target answer language
The corresponding product purchasing demand of sound data.Client sends target answer language according to the sentence pattern example provided in Products Show APP
Sound data, then corresponding product purchasing demand is got by target answer voice data, realize that intelligence gets client's
Product purchasing demand, to improve the speed for obtaining target product.It is complicated and changeable due to natural language, pass through natural language processing
Model handles target answer voice data, gets product purchasing demand compared to using modes such as keyword extractions,
Improve accuracy rate.
S30: product purchasing demand is matched with the attribute keywords in multidimensional data table, obtains the category of successful match
Property at least one corresponding product to be recommended of keyword.
Wherein, attribute keywords, which refer to, is stored in advance in multidimensional data table, keyword corresponding with configuring product, specifically
It can be age, gender, the requirements of support and insurance amount etc..Wherein, configuring product, which refers to, is pre-configured in multidimensional data table,
The product that can be recommended includes one or more configuring products in multidimensional data table.Product to be recommended refers to be bought with product
The corresponding product of demand.
Specifically, it is stored with multidimensional data table in database, includes the corresponding category of all configuring products in multidimensional data table
Property keyword, product purchasing demand is matched with each attribute keywords in multidimensional data table, if product purchasing demand with
Attribute keywords successful match, then using the corresponding configuring product of the attribute keywords of successful match as product to be recommended.Wherein,
It include at least one demand keyword in product purchasing demand, demand keyword quantity is more, accessed product to be recommended
It is more accurate.
S40: each product to be recommended matching degree corresponding with product purchasing demand is calculated.
Wherein, matching degree refers to the value that product to be recommended and product purchasing demand match, and matching degree is bigger, then to be recommended
Product is more matched with product purchasing demand, conversely, matching degree is smaller, then product to be recommended is more mismatched with product purchasing demand.
Specifically, database and server-side are connected to the network, and in the database, are previously stored with the table of comparisons, are wrapped in the table of comparisons
All configuring products are included, and are configured with corresponding weight for the corresponding attribute keywords of each configuring product, each attribute is crucial
Word is corresponding with multiple keywords, and the corresponding score value of each keyword is different.For example, attribute keywords are the requirements of support, ensureing is needed
It asks including several keywords such as " health insurances ", " casualty insurance ", " life insurance " and " juvenile's insurance ", and is each keyword
Configure corresponding score value.Firstly, determine corresponding with product purchasing demand product to be recommended, and in each product to be recommended of determination and
The attribute keywords of product purchasing demand successful match, and the table of comparisons is searched according to attribute keywords, it obtains and successful match
The corresponding score value of attribute keywords and preset weight, calculating each product to be recommended and product purchase by weighted formula needs
Seek corresponding matching degree.Wherein, weighted formula isY is matching degree, and n is attribute keywords quantity, AiIndicate i-th
The corresponding score value of a attribute keywords, wiIndicate the corresponding weight of ith attribute keyword.
Further, the corresponding table of comparisons of different demands type is different, analyzes product purchasing demand, determines client
The demand type of tendency, for example, demand type can be guarantee type and financing type, and type determines corresponding control according to demand
Table obtains weight corresponding with the attribute keywords of product purchasing demand successful match and score value, is calculated by weighted formula
Each product to be recommended matching degree corresponding with product purchasing demand.Wherein, each demand type is corresponding pre-configured
Weight and score value can be set according to actual needs, for example, demand type is guarantee type, then guarantee relevant to guarantee type
The corresponding weight such as amount is larger, so that it is more accurate finally to get matching degree.For another example demand type is financing type, that
The corresponding weights such as interest rate relevant to financing type are larger, so that root demand type determines the corresponding table of comparisons, and according to right
Corresponding weight and score are determined according to table, so that calculated matching is more accurate.
S50: being based on matching degree, descending sort is carried out at least one product to be recommended, by the production to be recommended of preset quantity
Product are determined as target product.
Wherein, target product refers to the biggish product of the matching chosen from product to be recommended.
Specifically, server-side obtains the corresponding matching degree of at least one product to be recommended, successively according to each production to be recommended
The size of the corresponding matching degree of product carries out descending sort, and the product to be recommended for obtaining preset quantity according to ranking results is determined as mesh
Product is marked, i.e. the acquisition biggish product to be recommended of matching degree is determined as target product.Wherein, present count is obtained according to ranking results
The product to be recommended of amount is determined as target product and specifically refers to according to ranking results, and selection comes top N (i.e. preset quantity)
Product to be recommended is determined as target product.Understandably, when the quantity of product to be recommended is not up to preset quantity, then will be needed
Recommended products is determined as target product.For example, preset quantity is 3, when the quantity of product to be recommended is 2, not up to present count
2 products to be recommended are determined as target product by amount.
S60: searching database based on each target product, obtain analysis data corresponding with each target product, will be every
One target product and corresponding analysis data feedback are to client.
Wherein, analysis data, which refer to, carries out similarities and differences point analysis institute according to pre-configured data corresponding with target product
The data of acquisition.Further, analysis data can be important clause corresponding with target product and feature.Wherein, similarities and differences
Analysis refers to, determines the corresponding important clause of each target product and feature, and extract the key of each important clause and feature
Word；Each target product important clause and the corresponding keyword of feature are mutually matched；By the keyword pair of successful match
The important clause or feature answered will match the corresponding important clause of unsuccessful keyword or feature as different as identical point
Point.For example, the corresponding keyword of the important clause of a certain target product is " public transport surprisingly die insurance money ", " 0-18 weeks
Year ", " dieing " and " 2 times of premiums of payment ", the corresponding keyword of feature are " insured amount high "；The corresponding important guarantor of another target product
The corresponding keyword of dangerous clause is " aviation casualty insurance gold ", " 18-30 one full year of life ", " great deformity " and " paying 1.5 times of premiums ",
The corresponding keyword of feature is " insured amount high ", and the corresponding important clause of two target products and the corresponding keyword of feature are carried out
Matching, can obtain, which is " insured amount high ", and difference is that important insurance clause is different.Wherein,
Protection amount refers to that the insurer undertakes reparation or pays the ceiling of insurance money responsibility.
Specifically, the corresponding data of each configuring product (specially important clause and feature) is stored in database.This
In embodiment, server obtains the quantity of all target products, and is stored in advance according in the quantity of target product and database
Determine that its analyzes data with the data of the target product.It wherein, will be with target product if the quantity of target product is 1
The analysis data and target product are fed back to client as analysis data by corresponding important clause and these data of feature.
Correspondingly, if the quantity of target product is at least two, database is searched according to each target product, is obtained and each target
The corresponding data of product, and similarities and differences point analysis and treated by similarities and differences point analysis is carried out to the corresponding data of each target product
As a result as analysis data corresponding with multiple target products, and analysis data and each target product are fed back into client.
Step S10-S60, the target corresponding with asked questions sent using Natural Language Processing Models to client are answered
Case voice data is analyzed and processed, and is realized and is carried out intellectual analysis to target answer voice data, is purchased with quick obtaining to product
Buy demand.Product purchasing demand is matched with the attribute keywords in multidimensional data table, to realize that quick obtaining largely accords with
Close the product to be recommended of product purchasing demand.By calculating each product to be recommended matching degree corresponding with product purchasing demand,
The higher product to be recommended of the matching degree of preset quantity is determined as target product, to improve the accuracy of target product recommendation.
Analysis data corresponding with target product are obtained, by each target product and corresponding analysis data feedback to client, client
Target product can be fully understanded according to analysis data, so that the important information for the target product that recommended products is recommended obtains body
It is existing.
In one embodiment, as shown in figure 3, before step S30, i.e., will be in product purchasing demand and multidimensional data table
Before the step of attribute keywords are matched, the Products Show method based on AI also specifically comprises the following steps:
S301: obtaining at least one products configuration information, each products configuration information include configuring product, logo,
Product attribute and form contract.
Wherein, products configuration information is the information for the progress products configuration that client is sent to server-side, products configuration letter
It include configuring product, logo, product attribute and form contract in breath.Wherein, refer to can in Products Show APP for configuring product
All products recommended.Logo refers to the corresponding mark of the affiliated company of configuring product.Product attribute refers to that company is pre-
First configured attribute corresponding with configuring product, for example, product attribute can be suitable year when configuring product is insurance products
Age is suitable for gender, the requirements of support and insurance amount etc..Form contract refers to that company has pre-established corresponding with configuring product
Contract, wherein include at least one clause in form contract.
S302: according to product attribute, text interpretation is carried out to each form contract, it is corresponding heavy to obtain each form contract
Want clause and feature.
Wherein, important clause refers to clause relatively important in form contract.It protects rights and interests for example, important clause can be
Clause corresponding with relief.Feature refers to configuring product content more outstanding, for example, be the advantages of a certain product " can rebating "
Deng.
Specifically, the corresponding important clause of different product attribute is different, according to product attribute, to the mark of each configuring product
Quasi contract is interpreted, and obtains the corresponding important clause of each configuring product and feature, and important clause, feature and configuration are produced
Product associated storage is in the database.More specifically, first determining that guarantee needs in the product attribute of each configuring product by expert
It asks, gets the important clause in each form contract further according to the requirements of support, and pass through the corresponding product of each configuring product
Attribute information, the characteristics of determining each configuring product.Wherein, the requirements of support can be " health insurance ", " casualty insurance ", " people
Longevity insurance " and " juvenile's insurance " etc..For example, a certain requirements of support are " health insurance ", then got from form contract
Important clause can be clause relevant to health, and if which disease can protect, which disease, which can not protect and settle a claim, compensates clause work
For important clause, then determine exemption clause as important clause etc. from standard insurance contract.For another example a certain configuring product pair
The product attribute information answered: Suitable Age is " the full 5 days babies that are born are teenage to 17 one full year of life ", the requirements of support are " two all risk insurances
Danger ", characteristic guarantee be " 31 kinds weight diseases protect, benefit this once get " etc., then it is determined that the configuring product the characteristics of can be " 31 kinds
Weight disease is protected, bonus return type ".
Further, newest contract corresponding with each configuring product is periodically crawled, and judges the newest conjunction that timing crawls
With whether identical as form contract；If newest contract is identical as form contract, be not required to the corresponding important clause of form contract and
Feature is updated；If newest contract and form contract be not identical, produced the newest newest contract got as the configuration
The corresponding form contract of product, and step S302 is executed, it is fixed to realize to update important clause corresponding with newest contract and feature
Shi Gengxin multidimensional data table, so that the subsequent analysis data got are more accurate.
S303: it is based on logo, product attribute, important clause and feature, obtains the corresponding attribute of each configuring product
Keyword.
Wherein, attribute keywords refer to keyword corresponding with configuring product.Specifically, it is corresponding to obtain each configuring product
Logo, product attribute, important clause and feature, the keyword in important clause is determined, by logo, product category
Property, the corresponding keyword of important clause and feature are as attribute keywords corresponding with the configuring product.For example, attribute keywords
It can be Business Name, name of product, age, gender, the requirements of support, characteristic guarantee, insurance amount, feature and important clause pair
The keyword etc. answered.
S304: being based at least one configuring product and corresponding attribute keywords, constructs multidimensional data table.
Specifically, each configuring product and corresponding attribute keywords are associated, construct multi-dimensional data table, multidimensional
Tables of data is the tables of data based on the building of the dimensions such as logo, product attribute, important clause and feature.By constructing multidimensional number
According to table, product to be recommended is got from multi-dimensional data table so as to subsequent.
Step S301-S304, obtains at least one products configuration information, each products configuration information include configuring product,
Logo, product attribute and form contract configure products configuration information so as to subsequent.According to product attribute, to every
One form contract carries out text interpretation, obtains the corresponding important clause of each form contract and feature to realize, is subsequent acquisition
Product to be recommended and analysis data provide technical support.Based on logo, product attribute, important clause and feature, obtain every
The corresponding attribute keywords of one configuring product, to get product to be recommended according to product purchasing demand and attribute keywords.
Based at least one configuring product and corresponding attribute keywords, multidimensional data table is constructed, realization is pre-configured with multidimensional data table,
It is analyzed without treating recommended products in Products Show is crossed into, to improve Products Show rate.
In one embodiment, product purchasing demand include at least one demand keyword, wherein demand keyword refer to from
The keyword extracted in target answer voice data can get product to be recommended by demand keyword.
As shown in figure 4, product purchasing demand is matched with the attribute keywords in multidimensional data table, obtain matching at
At least one corresponding product to be recommended of the attribute keywords of function, specifically comprises the following steps:
S31: at least one demand keyword attribute keywords corresponding with multidimensional data table are matched.
It wherein, include the corresponding attribute keywords of each configuring product in multidimensional data table.In the present embodiment, server-side will
At least one demand keyword is matched with the attribute keywords in multidimensional data table, in particular to by by demand keyword
Attribute keywords corresponding with configuring product each in multidimensional data table are matched, so that subsequent realization is from multidimensional data table
Determine product to be recommended.
S32:, will matching if at least one demand keyword attribute keywords successful match corresponding with multidimensional data table
Successful attribute keywords will include at least one configuring product of target keyword in multidimensional data table as target keyword
It is determined as product to be recommended.
Wherein, target keyword refers to the keyword to match with demand keyword obtained from multidimensional data table.
Specifically, server-side extracts at least one demand keyword from target answer voice data, if at least one is needed
The attribute keywords successful match in keyword and multidimensional data table is sought, using the attribute keywords of successful match as target critical
Word, and at least one configuring product in multidimensional data table comprising target keyword is determined as product to be recommended.For example, product
The demand keyword for including in purchasing demand B is " life insurance " and " 30-40 years old ", attribute keywords including " life insurance " and
" 30-50 years old ", by matching demand keyword with the attribute keywords in multidimensional data table, if in multidimensional data table
Attribute keywords and demand keyword are " life insurance " and " 30-40 years old " successful match, then will be " life insurance with demand keyword
Insurance " and " 30-40 years old " corresponding attribute keywords " life insurance " and " 30-40 years old " are determined as target keyword, and will be more
Comprising " life insurance " and " 30-40 years old " at least one corresponding configuring product as product to be recommended in dimension data table.
S33: if the attribute keywords matching corresponding with multidimensional data table of all demand keywords is unsuccessful, to client
Send the information for reacquiring target answer voice data.
Specifically, if attribute closes from all demand keywords and multidimensional data table extracted in target answer voice data
Keyword match it is unsuccessful, then not comprising the matching product that matches with demand keyword from multidimensional data table, then to client
End sends the information for reacquiring target answer voice data.
In step S31-S33, if the matching of at least one demand keyword corresponding with multidimensional data table attribute keywords at
Function, then using the attribute keywords of successful match as target keyword, will in multidimensional data table comprising target keyword at least
One configuring product is determined as product to be recommended, and to realize quick obtaining to product to be recommended, acquisition methods are simple and quick.If institute
There is the attribute keywords matching corresponding with multidimensional data table of demand keyword unsuccessful, is then sent to client and reacquire target
The information of answer voice data, to determine product to be recommended according to target answer voice data again, so that product to be recommended
Accuracy is high.
In one embodiment, as shown in figure 5, step S40, that is, it is corresponding with product purchasing demand to calculate each product to be recommended
Matching degree, specifically comprise the following steps:
S41: it is based on each product to be recommended, obtains at least one corresponding target keyword of product to be recommended.
Specifically, server-side obtains each product to be recommended, and each corresponding one or more attributes of product to be recommended are crucial
Word obtains the corresponding attribute keywords corresponding with demand keyword of each product to be recommended, that is, gets and product to be recommended
Corresponding target keyword, so as to each product to be recommended of subsequent determination matching degree corresponding with product purchasing demand.
S42: searching database according to target keyword, obtains default weight corresponding with each target keyword and divides
Value.
Specifically, configured with corresponding weight and divide for the corresponding attribute keywords of each configuring product in advance in database
Value, wherein the corresponding target keyword of each product to be recommended is one or more.Database is searched by target keyword,
Obtain default weight corresponding with each target keyword and score value.For example, the requirements of support include " health insurance ", " unexpected guarantor
Several keywords such as danger ", " life insurance " and " juvenile's insurance ", target keyword are " casualty insurance ", then obtaining and " guarantee
The corresponding weight of demand ", and obtain " casualty insurance " corresponding score value.
S43: being based on the corresponding default weight of each target keyword and score value, is calculated using weighted formula each to be recommended
The corresponding matching degree of product.
Specifically, server-side determines the corresponding target keyword of each product to be recommended, corresponding default weight and score value,
Matching degree corresponding with each product to be recommended is calculated by weighted formula.Wherein, weighted formula isY is
With degree, n is target keyword quantity, AiIndicate the corresponding score value of i-th of target keyword, wiIndicate i-th of target keyword pair
The weight answered.For example, the corresponding target keyword of a certain product to be recommended is " gender is women ", " requirements of support are accident insurance "
" insurance amount be 200,000 " obtains and gender, the requirements of support and insures the corresponding default weight of amount, and obtain " women ",
" accident insurance " and " 200,000 " corresponding score value, and the corresponding matching degree of the product to be recommended is calculated using weighted formula.It can manage
Xie Di, the target keyword got is more, and the corresponding weight of target keyword is bigger, i.e., the target of the product to be recommended is closed
The number of keyword and demand Keywords matching in product purchasing demand is more, and with the successful target critical of demand Keywords matching
Word is important keyword, then matching degree is bigger.
In step S41-S43, it is based on each product to be recommended, obtains at least one corresponding target critical of product to be recommended
Word, to determine the corresponding matching degree of recommended products to be determined according to target keyword.It is corresponding based on each target keyword
Default weight and score value calculate the corresponding matching degree of each product to be recommended using weighted formula, realize that determination is each to be recommended
Matching degree between product and product purchasing demand, and determine that target product provides technical support according to matching degree to be subsequent, with
Improve the accuracy of Products Show.
In one embodiment, as shown in fig. 6, in step S60, database is searched based on each target product, is obtained and every
The corresponding analysis data of one target product, specifically comprise the following steps:
S61: database is searched based on each target product, obtains important clause corresponding with each target product and spy
Point.
Specifically, the incidence relation of configuring product, important clause and feature, i.e. step S302 are previously stored in database
In, according to product attribute, each configuring product has been got in advance and has corresponded to important clause and feature.Server-side passes through target product
Database is searched, important clause corresponding with each target product and feature can be got.It is subsequent to be corresponded to according to target product
Important clause and feature, determine corresponding with target product analysis data.
S62: being based on multidimensional data table, obtains corresponding with each important clause and feature attribute keywords, as with mesh
Mark the corresponding keyword of product.
Specifically, the keyword that server first passes through that keyword extraction algorithm extracts in important clause and feature in advance is used as category
Property keyword.Specifically, server-side is first based on each important clause and feature, searches multidimensional data table, obtain with it is each important
Clause and the corresponding attribute keywords of feature, as keyword corresponding with target product.Attribute keywords refer to be produced with target
The corresponding keyword of product, keyword specifically corresponding with the important clause of the target product and feature.
S63: similarities and differences point analysis is carried out to the corresponding keyword of each target product, is obtained corresponding with each target product
Analyze data.
Specifically, the corresponding keyword of each target product is mutually matched；The keyword of successful match is corresponding
Important clause or feature as identical point, the corresponding important clause of unsuccessful keyword or feature will be matched as different
Point；Pre-set analysis template is obtained, each target product, identical point and difference are filled into analysis template, with
Get analysis data corresponding with target product, and by analyze in the form of template will analysis data feedback to client.More specifically
Ground determines all keywords that each important clause and each feature include in each target product, by the keyword and remaining
All keywords that the corresponding each important clause of target product and each feature include are matched, if a certain target product
All keywords that one important clause includes, all keywords for including with an important clause of another target product are matched into
Function, then the important clause belongs to common ground.
It further, will important clause corresponding with target product and spy if the target product got is one
Point is filled directly into analysis template, to get analysis data corresponding with target product, and by analyze in the form of template will point
Data feedback is analysed to client.
In step S61-S63, database is searched based on each target product, is obtained corresponding with each target product important
Clause and feature can also feed back important clause and feature when so as to the subsequent target product to client feedback.Based on multidimensional data
Table obtains attribute keywords corresponding with each important clause and feature, as keyword corresponding with target product, to realize
The keyword for obtaining target product, provides technical support for subsequent similarities and differences point analysis.Keyword corresponding to each target product
Similarities and differences point analysis is carried out, analysis data corresponding with each target product are obtained, it is identical between getting target product to realize
Point and difference, so that client becomes more apparent upon target product, so that the important information of target product is embodied.
In one embodiment, as shown in fig. 7, after step S60, i.e., by each target product and corresponding analysis number
After the step of feeding back to client, the Products Show method based on AI also specifically comprises the following steps:
S601: obtaining the request that places an order that client is sent, and the request that places an order includes target product.
Specifically, server-side to client feedback target product corresponding with target answer voice data and analyzes data,
Wherein, also to client feedback lower singular link corresponding with each target product, client can be placed an order based on the transmission of lower singular link and be asked
It asks, comprising target product in the request that places an order.
S602: inquiring database based on target product, obtains speech sound eeplaining recording corresponding with target product.
Wherein, speech sound eeplaining recording refers to important clause corresponding with target product, and line by line by speech sound eeplaining mode
It records to each important clause.Specifically, it is stored with the corresponding speech sound eeplaining recording of each target product in database, leads to
It crosses target product and searches database, speech sound eeplaining recording corresponding with target product is obtained, so as to subsequent for client playing voice
Explanation recording discharges manpower without manually explaining for client.
S603: by recording for client terminal playing speech sound eeplaining, double record operation datas of client feedback are obtained, to complete
Lower single operation.
Specifically, the explanation comprising each important clause corresponding with target product is recorded in speech sound eeplaining recording, wherein
The voice data also whether having appreciated that comprising inquiry client in speech sound eeplaining recording, specifically can be, and play each weight
The voice data whether inquiry client has appreciated that is intercutted after wanting clause.When server-side by speech sound eeplaining recording broadcasting to client
End obtains double record operation datas of client feedback in real time, wherein double record operation datas are synchronized video recording and recording data.It can
To understand ground, based on the voice data that whether has appreciated that of inquiry client that server-side is sent, when getting what client was sent
After the data having appreciated that, then speech sound eeplaining is carried out to next important clause.
Step S601-S603 obtains the request that places an order that client is sent, and the request that places an order includes target product, is based on target
Product inquires database, speech sound eeplaining recording corresponding with target product is obtained, so as to subsequently through speech sound eeplaining recording pair
Target product is explained, without discharging manpower by manually explaining to target product.By for client terminal playing voice
Explanation recording, obtains double record operation datas of client feedback, completes lower single operation to realize, improves single efficiency under product.
In one embodiment, the corresponding customer ID of target answer voice data, wherein customer ID has uniqueness,
Unique client can be found by the customer ID, and client properties information corresponding with the client can be got.
After step S60, i.e., by each target product and corresponding analysis data feedback to client the step of after,
Products Show method based on AI further includes following steps:
(1) client properties information corresponding with customer ID is obtained according to customer ID.
Specifically, when to client feedback target product and analysis data, detect whether the client has logged in product
Recommend APP, if having logged in, directly by each target product and corresponding analysis data feedback to client, if being not logged in,
The information that need to be logged in client feedback, and determine that client corresponding with target answer voice data marks by user login information
Know, database is searched by the customer ID, obtains client properties information corresponding with customer ID, wherein client properties letter
Breath refers to storage customer information in the database, client properties information can be the age corresponding with customer ID, gender,
Domain, commercial circle, occupation and marital status etc. are drawn a portrait so as to subsequent according to client properties information architecture user.
(2) at least one corresponding demand keyword of client properties information and product purchasing demand is associated, is constructed
User's portrait.
Specifically, it after server-side gets the corresponding client properties information of customer ID, obtains from target answer voice number
At least one the demand keyword extracted in can be got and client properties information phase by least one demand keyword
It is corresponding that the preference of product and demand etc. are drawn based on client properties information and at least one demand keyword with constructing user
Picture, and save into database.If subsequent have new user to log in Products Show APP, and is not based on asked questions transmission target and answers
Case voice data can draw a portrait based in the corresponding client properties information searching database of new user according to the user in database
Demand keyword corresponding with the client properties information is obtained, keyword is new user's recommended products according to demand, and will be recommended
Product is shown in the display page corresponding with new user, to improve Products Show efficiency.
Further, as in step S33, if the attribute keywords matching corresponding with multidimensional data table of all demand keywords
It is unsuccessful, then can get client properties information by keyword according to demand, according to client properties information searching database, obtain with
The corresponding user's portrait of the client properties information, is drawn a portrait by the user and determines target requirement corresponding with the client properties information
Keyword is client feedback recommended products corresponding with target requirement keyword.
It should be understood that the size of the serial number of each step is not meant that the order of the execution order in above-described embodiment, each process
Execution sequence should be determined by its function and internal logic, the implementation process without coping with the embodiment of the present invention constitutes any limit
It is fixed.
In one embodiment, a kind of Products Show device based on AI is provided, should Products Show device based on AI with it is upper
The Products Show method in embodiment based on AI is stated to correspond.As shown in figure 8, should include answering based on the Products Show device of AI
Case obtains module 10, requirement extract module 20, product to be recommended and obtains module 30, matching degree computing module 40, target product and obtain
Modulus block 50 and data feedback module 60.
Detailed description are as follows for each functional module:
Answer obtains module 10, the target answer voice data sent for obtaining client based on asked questions.
Requirement extract module 20, for being carried out at analysis using Natural Language Processing Models to target answer voice data
Reason obtains product purchasing demand corresponding with target answer voice data.
Product to be recommended obtains module 30, for carrying out the attribute keywords in product purchasing demand and multidimensional data table
Matching, obtains at least one corresponding product to be recommended of attribute keywords of successful match.
Matching degree computing module 40, for calculating each product to be recommended matching degree corresponding with product purchasing demand.
Target product obtains module 50, for being based on matching degree, carries out descending sort at least one product to be recommended, will
The product to be recommended of preset quantity is determined as target product.
Data feedback module 60 obtains corresponding with each target product for searching database based on each target product
Analysis data, by each target product and corresponding analysis data feedback to client.
In one embodiment, before product to be recommended obtains module 30, the Products Show device based on AI, further includes matching
Set information acquisition unit, analytical unit, attribute keywords determination unit and multidimensional data table construction unit.
Configuration information acquiring unit, for obtaining at least one products configuration information, each products configuration information includes matching
Set product, logo, product attribute and form contract.
Analytical unit, for carrying out text interpretation to each form contract, obtaining each form contract according to product attribute
Corresponding important clause and feature.
Attribute keywords determination unit obtains each for being based on logo, product attribute, important clause and feature
The corresponding attribute keywords of configuring product.
Multidimensional data table construction unit constructs more for being based at least one configuring product and corresponding attribute keywords
Dimension data table.
In one embodiment, product purchasing demand includes at least one demand keyword.
Product to be recommended obtains module 30, including matching unit, the first matching determination unit and the second matching determination unit.
Matching unit is used for the attribute keywords progress corresponding with multidimensional data table of at least one demand keyword
Match.
First matching determination unit, if being used at least one demand keyword attribute keywords corresponding with multidimensional data table
Successful match will include target keyword in multidimensional data table then using the attribute keywords of successful match as target keyword
At least one configuring product be determined as product to be recommended.
Second matching determination unit, if for the attribute keywords matching corresponding with multidimensional data table of all demand keywords
It is unsuccessful, then the information for reacquiring target answer voice data is sent to client.
In one embodiment, matching degree computing module 40, including target keyword acquiring unit, parameter acquisition module and
With degree computing module.
Target keyword acquiring unit obtains product to be recommended corresponding at least one for being based on each product to be recommended
A target keyword.
Parameter acquisition module obtains corresponding with each target keyword for searching database according to target keyword
Default weight and score value.
Matching degree computing module, it is public using weighting for being based on the corresponding default weight of each target keyword and score value
Formula calculates the corresponding matching degree of each product to be recommended.
In one embodiment, data feedback module 60, including important clause and feature acquiring unit, keyword determination unit
With analysis data determination unit.
Important clause and feature acquiring unit obtain and each target for searching database based on each target product
The corresponding important clause of product and feature.
Keyword determination unit obtains attribute corresponding with each important clause and feature for being based on multidimensional data table
Keyword, as keyword corresponding with target product.
Analyze data determination unit, for the corresponding keyword of each target product carry out similarities and differences point analysis, obtain with
The corresponding analysis data of each target product.
In one embodiment, after data feedback module 60, the Products Show device based on AI further includes request
Unit, explanation recording acquiring unit and lower single operating unit.
Request unit, for obtaining the request that places an order of client transmission, the request that places an order includes target product.
Explanation recording acquiring unit obtains language corresponding with target product for inquiring database based on target product
Sound explanation recording.
Lower list operating unit, for by the way that for the recording of client terminal playing speech sound eeplaining, the double records for obtaining client feedback are grasped
Make data, to complete lower single operation.
In one embodiment, the corresponding customer ID of target answer voice data.
After data feedback module 60, the Products Show device based on AI further include client properties information acquisition unit and
User's portrait construction unit.
Client properties information acquisition unit is believed for obtaining client properties corresponding with customer ID according to customer ID
Breath.
User's portrait construction unit, for closing client properties information and at least one corresponding demand of product purchasing demand
Keyword is associated, building user's portrait.
Specific restriction about the Products Show device based on AI may refer to above for the Products Show based on AI
The restriction of method, details are not described herein.Modules in the above-mentioned Products Show device based on AI can be fully or partially through
Software, hardware and its group and to realize.Above-mentioned each module can be embedded in the form of hardware or independently of the place in computer equipment
It manages in device, can also be stored in a software form in the memory in computer equipment, in order to which processor calls execution or more
The corresponding operation of modules.
In one embodiment, a kind of computer equipment is provided, which can be server-side, internal junction
Composition can be as shown in Figure 9.The computer equipment include by system bus connect processor, memory, network interface and
Database.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory packet of the computer equipment
Include non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system, computer program and data
Library.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculating
The database of machine equipment is for analyzing data, multidimensional data table and the corresponding weight of each attribute keywords and score value etc..The meter
The network interface for calculating machine equipment is used to communicate with external terminal by network connection.When the computer program is executed by processor
To realize a kind of Products Show method based on AI.
In one embodiment, a kind of computer equipment is provided, including memory, processor and storage are on a memory simultaneously
The computer program that can be run on a processor, processor realize the production in above-described embodiment based on AI when executing computer program
The step of product recommended method, for example, step S10 shown in Fig. 2 to step S60, alternatively, such as Fig. 3 to step shown in Fig. 7.Place
Reason device realizes the function of each module in the Products Show device in above-described embodiment based on AI, example when executing computer program
Such as, module 10 shown in Fig. 8 to module 60 function.To avoid repeating, details are not described herein again.
In one embodiment, a kind of computer readable storage medium is provided, computer program, computer are stored thereon with
The Products Show method based on AI in above method embodiment is realized when program is executed by processor, for example, step shown in Fig. 2
S10 to step S60, alternatively, such as Fig. 3 to step shown in Fig. 7.The computer program realizes above-mentioned implementation when being executed by processor
In example in the Products Show device based on AI each module function, for example, function of the module 10 shown in Fig. 8 to module 60.To keep away
Exempt to repeat, details are not described herein again.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, computer program can be stored in a non-volatile computer and can be read
In storage medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein, the application
To any reference of memory, storage, database or other media used in provided each embodiment, may each comprise non-
Volatibility and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM (PROM),
Electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include arbitrary access
Memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms, such as static
RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhanced SDRAM
(ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (RambuS) directly RAM (RDRAM), straight
Connect memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
It is apparent to those skilled in the art that for convenience of description and succinctly, only with above-mentioned each function
Can unit, module division progress for example, in practical application, can according to need and by above-mentioned function distribution by different
Functional unit, module are completed, i.e., the internal structure of device are divided into different functional unit or module, to complete above description
All or part of function.
The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although with reference to the foregoing embodiments
Invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each implementation
Technical solution documented by example is modified or equivalent replacement of some of the technical features；And these modification or
Replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution should all include
Within protection scope of the present invention.
Claims (10)
1. a kind of Products Show method based on AI characterized by comprising
Obtain the target answer voice data that client is sent based on asked questions；
The target answer voice data is analyzed and processed using Natural Language Processing Models, is obtained and the target answer
The corresponding product purchasing demand of voice data；
The product purchasing demand is matched with the attribute keywords in multidimensional data table, the attribute for obtaining successful match closes
At least one corresponding product to be recommended of keyword；
Calculate each product to be recommended matching degree corresponding with the product purchasing demand；
Based on the matching degree, descending sort is carried out at least one described product to be recommended, by described in preset quantity wait push away
It recommends product and is determined as target product；
Database is searched based on each target product, obtains analysis data corresponding with each target product, it will be every
One target product and the corresponding analysis data feedback give the client.
2. the Products Show method based on AI as described in claim 1, which is characterized in that needed in described buy the product
Before asking the step of being matched with the attribute keywords in multidimensional data table, the Products Show method based on AI is also wrapped
It includes:
At least one products configuration information is obtained, each products configuration information includes configuring product, logo, product category
Property and form contract；
According to the product attribute, text interpretation is carried out to each form contract, it is corresponding to obtain each form contract
Important clause and feature；
Based on the logo, the product attribute, the important clause and the feature, each configuring product is obtained
Corresponding attribute keywords；
Based at least one configuring product and corresponding attribute keywords, multidimensional data table is constructed.
3. the Products Show method based on AI as claimed in claim 2, which is characterized in that the product purchasing demand includes extremely
A few demand keyword；
It is described to match the product purchasing demand with the attribute keywords in multidimensional data table, obtain the category of successful match
Property at least one corresponding product to be recommended of keyword, comprising:
At least one described demand keyword attribute keywords corresponding with the multidimensional data table are matched；
It, will matching if at least one described demand keyword attribute keywords successful match corresponding with the multidimensional data table
The successful attribute keywords are used as target keyword, will in the multidimensional data table comprising the target keyword at least
One configuring product is determined as product to be recommended；
If the attribute keywords matching corresponding with the multidimensional data table of all demand keywords is unsuccessful, to the visitor
Family end sends the information for reacquiring target answer voice data.
4. the Products Show method based on AI as claimed in claim 3, which is characterized in that described to be calculated often using weighted formula
One product to be recommended matching degree corresponding with the product purchasing demand, comprising:
Based on each product to be recommended, at least one corresponding target keyword of the product to be recommended is obtained；
Database is searched according to the target keyword, obtain default weight corresponding with each target keyword and is divided
Value；
Based on each corresponding default weight of target keyword and the score value, each institute is calculated using weighted formula
State the corresponding matching degree of product to be recommended.
5. the Products Show method based on AI as claimed in claim 2, which is characterized in that described to be produced based on each target
Product search database, obtain analysis data corresponding with each target product, comprising:
Database is searched based on each target product, obtains important clause corresponding with each target product and spy
Point；
Based on the multidimensional data table, attribute keywords corresponding with each important clause and the feature are obtained, as
Keyword corresponding with the target product；
Similarities and differences point analysis is carried out to the corresponding keyword of each target product, is obtained corresponding with each target product
Analyze data.
6. the Products Show method based on AI as described in claim 1, which is characterized in that produce each target described
After the step of product and the corresponding analysis data feedback give the client, the Products Show method based on AI is also wrapped
It includes:
The request that places an order that client is sent is obtained, the request that places an order includes target product；
Database is inquired based on the target product, obtains speech sound eeplaining recording corresponding with the target product；
By for speech sound eeplaining described in the client terminal playing recording, obtain double record operation datas of the client feedback, with
Complete lower single operation.
7. the Products Show method based on AI as claimed in claim 3, which is characterized in that the target answer voice data pair
Answer a customer ID；
After described the step of giving each target product and the corresponding analysis data feedback to the client, institute
State the Products Show method based on AI further include:
Client properties information corresponding with the customer ID is obtained according to the customer ID；
The client properties information and at least one corresponding demand keyword of the product purchasing demand are associated, constructed
User's portrait.
8. a kind of Products Show device based on AI characterized by comprising
Answer obtains module, the target answer voice data sent for obtaining client based on asked questions；
Requirement extract module, for being analyzed and processed using Natural Language Processing Models to the target answer voice data,
Obtain product purchasing demand corresponding with the target answer voice data；
Product to be recommended obtains module, for carrying out the attribute keywords in the product purchasing demand and multidimensional data table
Match, obtains at least one corresponding product to be recommended of attribute keywords of successful match；
Matching degree computing module, for calculating each product to be recommended matching degree corresponding with the product purchasing demand；
Target product obtains module, for being based on the matching degree, carries out descending sort at least one described product to be recommended,
The product to be recommended of preset quantity is determined as target product；
Data feedback module obtains and each target product pair for searching database based on each target product
The analysis data answered give each target product and the corresponding analysis data feedback to the client.
9. a kind of computer equipment, including memory, processor and storage are in the memory and can be in the processor
The computer program of upper operation, which is characterized in that the processor realized when executing the computer program as claim 1 to
The step of Products Show method described in 7 any one based on AI.
10. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, and feature exists
In realization Products Show side based on AI as described in any one of claim 1 to 7 when the computer program is executed by processor
The step of method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811537638.3A CN109727092A (en) | 2018-12-15 | 2018-12-15 | Products Show method, apparatus, computer equipment and storage medium based on AI |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811537638.3A CN109727092A (en) | 2018-12-15 | 2018-12-15 | Products Show method, apparatus, computer equipment and storage medium based on AI |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109727092A true CN109727092A (en) | 2019-05-07 |
Family
ID=66296114
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811537638.3A Pending CN109727092A (en) | 2018-12-15 | 2018-12-15 | Products Show method, apparatus, computer equipment and storage medium based on AI |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109727092A (en) |
Cited By (32)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110163723A (en) * | 2019-05-20 | 2019-08-23 | 深圳市和讯华谷信息技术有限公司 | Recommended method, device, computer equipment and storage medium based on product feature |
| CN110246494A (en) * | 2019-05-20 | 2019-09-17 | 深圳壹账通智能科技有限公司 | Service request method, device and computer equipment based on speech recognition |
| CN110297968A (en) * | 2019-05-22 | 2019-10-01 | 中国平安人寿保险股份有限公司 | Product method for pushing, device, computer equipment and storage medium |
| CN110400180A (en) * | 2019-07-29 | 2019-11-01 | 腾讯科技（深圳）有限公司 | Display methods, device and storage medium based on recommendation information |
| CN110675871A (en) * | 2019-09-25 | 2020-01-10 | 北京蓦然认知科技有限公司 | Voice recognition method and device |
| CN111091832A (en) * | 2019-11-28 | 2020-05-01 | 秒针信息技术有限公司 | Intention assessment method and system based on voice recognition |
| CN111160989A (en) * | 2019-12-31 | 2020-05-15 | 中国银行股份有限公司 | Information processing method and device |
| CN111311353A (en) * | 2020-01-19 | 2020-06-19 | 河北工业职业技术学院 | User purchase intention evaluation method and system and terminal equipment |
| CN111539221A (en) * | 2020-05-13 | 2020-08-14 | 北京焦点新干线信息技术有限公司 | Data processing method and system |
| CN111626813A (en) * | 2020-04-22 | 2020-09-04 | 北京健康之家科技有限公司 | Product recommendation method and system |
| CN112100491A (en) * | 2020-09-08 | 2020-12-18 | 中国平安人寿保险股份有限公司 | Information recommendation method, device and equipment based on user data and storage medium |
| CN112148958A (en) * | 2019-06-26 | 2020-12-29 | 傲基科技股份有限公司 | Method, apparatus, and computer storage medium for information recommendation |
| CN112182394A (en) * | 2020-10-10 | 2021-01-05 | 深圳壹账通智能科技有限公司 | Product recommendation method and device, electronic equipment and storage medium |
| CN112418996A (en) * | 2020-11-30 | 2021-02-26 | 珠海采筑电子商务有限公司 | Recommendation method and system for elevator suppliers |
| CN112465592A (en) * | 2020-11-26 | 2021-03-09 | 珠海新海通电子商务有限公司 | Intelligent commodity recommending method based on platform |
| CN112488750A (en) * | 2020-11-27 | 2021-03-12 | 上海容大数字技术有限公司 | Intelligent recommendation and renewal system for insurance scene |
| CN112633992A (en) * | 2021-01-11 | 2021-04-09 | 上海明略人工智能(集团)有限公司 | Sales management method and system based on voice recognition |
| CN112632277A (en) * | 2020-12-15 | 2021-04-09 | 五八同城信息技术有限公司 | Resource processing method and device for target content object |
| CN112651862A (en) * | 2020-12-24 | 2021-04-13 | 成都存时科技有限公司 | Student academic development direction planning method, device, equipment and readable storage medium |
| CN112861005A (en) * | 2021-02-22 | 2021-05-28 | 上海风秩科技有限公司 | Method, device and equipment for information push |
| CN112950326A (en) * | 2021-03-17 | 2021-06-11 | 南通大学 | Artificial intelligence data analysis system supporting deep learning working principle |
| CN112967721A (en) * | 2021-02-03 | 2021-06-15 | 上海明略人工智能(集团)有限公司 | Sales lead information identification method and system based on voice identification technology |
| CN113282783A (en) * | 2021-05-28 | 2021-08-20 | 上海明略人工智能(集团)有限公司 | Method and device for information recommendation, electronic equipment and readable storage medium |
| WO2021196708A1 (en) * | 2020-03-31 | 2021-10-07 | 北京大米科技有限公司 | Method and apparatus for online interaction, storage medium and electronic device |
| CN113538048A (en) * | 2021-07-12 | 2021-10-22 | 深圳市明源云客电子商务有限公司 | Demand information obtaining method and device, terminal equipment and storage medium |
| CN113593552A (en) * | 2021-07-12 | 2021-11-02 | 深圳市明源云客电子商务有限公司 | Demand information statistical method and device, terminal equipment and storage medium |
| CN113836379A (en) * | 2021-09-26 | 2021-12-24 | 北京百炼智能科技有限公司 | Intelligent recommendation method and system based on customer image |
| CN114866818A (en) * | 2022-06-17 | 2022-08-05 | 深圳壹账通智能科技有限公司 | Video recommendation method and device, computer equipment and storage medium |
| CN116303975A (en) * | 2023-05-11 | 2023-06-23 | 腾讯科技（深圳）有限公司 | Training method of recall model, recall method and related equipment |
| CN117312878A (en) * | 2023-10-30 | 2023-12-29 | 青岛汇聚融合健康科技有限公司 | Method, system and device for generating mall user mark |
| CN117710122A (en) * | 2024-01-11 | 2024-03-15 | 广州小锤科技服务有限公司 | Big data claim product management method, device, computer equipment and storage medium |
| CN112967721B (en) * | 2021-02-03 | 2024-05-31 | 上海明略人工智能(集团)有限公司 | Sales lead information recognition method and system based on voice recognition technology |
-
2018
- 2018-12-15 CN CN201811537638.3A patent/CN109727092A/en active Pending
Cited By (45)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110246494A (en) * | 2019-05-20 | 2019-09-17 | 深圳壹账通智能科技有限公司 | Service request method, device and computer equipment based on speech recognition |
| CN110163723A (en) * | 2019-05-20 | 2019-08-23 | 深圳市和讯华谷信息技术有限公司 | Recommended method, device, computer equipment and storage medium based on product feature |
| CN110297968A (en) * | 2019-05-22 | 2019-10-01 | 中国平安人寿保险股份有限公司 | Product method for pushing, device, computer equipment and storage medium |
| CN110297968B (en) * | 2019-05-22 | 2023-10-31 | 中国平安人寿保险股份有限公司 | Product pushing method, device, computer equipment and storage medium |
| CN112148958A (en) * | 2019-06-26 | 2020-12-29 | 傲基科技股份有限公司 | Method, apparatus, and computer storage medium for information recommendation |
| CN110400180A (en) * | 2019-07-29 | 2019-11-01 | 腾讯科技（深圳）有限公司 | Display methods, device and storage medium based on recommendation information |
| CN110400180B (en) * | 2019-07-29 | 2023-11-07 | 腾讯科技（深圳）有限公司 | Recommendation information-based display method and device and storage medium |
| CN110675871B (en) * | 2019-09-25 | 2021-03-05 | 北京蓦然认知科技有限公司 | Voice recognition method and device |
| CN110675871A (en) * | 2019-09-25 | 2020-01-10 | 北京蓦然认知科技有限公司 | Voice recognition method and device |
| CN111091832B (en) * | 2019-11-28 | 2022-12-30 | 秒针信息技术有限公司 | Intention assessment method and system based on voice recognition |
| CN111091832A (en) * | 2019-11-28 | 2020-05-01 | 秒针信息技术有限公司 | Intention assessment method and system based on voice recognition |
| CN111160989B (en) * | 2019-12-31 | 2023-08-08 | 中国银行股份有限公司 | Information processing method and device |
| CN111160989A (en) * | 2019-12-31 | 2020-05-15 | 中国银行股份有限公司 | Information processing method and device |
| CN111311353A (en) * | 2020-01-19 | 2020-06-19 | 河北工业职业技术学院 | User purchase intention evaluation method and system and terminal equipment |
| CN111311353B (en) * | 2020-01-19 | 2023-06-30 | 河北工业职业技术学院 | User purchase intention evaluation method, system and terminal equipment |
| WO2021196708A1 (en) * | 2020-03-31 | 2021-10-07 | 北京大米科技有限公司 | Method and apparatus for online interaction, storage medium and electronic device |
| CN111626813B (en) * | 2020-04-22 | 2023-09-29 | 北京水滴科技集团有限公司 | Product recommendation method and system |
| CN111626813A (en) * | 2020-04-22 | 2020-09-04 | 北京健康之家科技有限公司 | Product recommendation method and system |
| CN111539221B (en) * | 2020-05-13 | 2023-09-12 | 北京焦点新干线信息技术有限公司 | Data processing method and system |
| CN111539221A (en) * | 2020-05-13 | 2020-08-14 | 北京焦点新干线信息技术有限公司 | Data processing method and system |
| CN112100491A (en) * | 2020-09-08 | 2020-12-18 | 中国平安人寿保险股份有限公司 | Information recommendation method, device and equipment based on user data and storage medium |
| CN112182394A (en) * | 2020-10-10 | 2021-01-05 | 深圳壹账通智能科技有限公司 | Product recommendation method and device, electronic equipment and storage medium |
| WO2022073517A1 (en) * | 2020-10-10 | 2022-04-14 | 深圳壹账通智能科技有限公司 | Product recommendation method and apparatus, and electronic device and storage medium |
| CN112465592A (en) * | 2020-11-26 | 2021-03-09 | 珠海新海通电子商务有限公司 | Intelligent commodity recommending method based on platform |
| CN112488750A (en) * | 2020-11-27 | 2021-03-12 | 上海容大数字技术有限公司 | Intelligent recommendation and renewal system for insurance scene |
| CN112418996A (en) * | 2020-11-30 | 2021-02-26 | 珠海采筑电子商务有限公司 | Recommendation method and system for elevator suppliers |
| CN112632277A (en) * | 2020-12-15 | 2021-04-09 | 五八同城信息技术有限公司 | Resource processing method and device for target content object |
| CN112651862A (en) * | 2020-12-24 | 2021-04-13 | 成都存时科技有限公司 | Student academic development direction planning method, device, equipment and readable storage medium |
| CN112633992A (en) * | 2021-01-11 | 2021-04-09 | 上海明略人工智能(集团)有限公司 | Sales management method and system based on voice recognition |
| CN112967721A (en) * | 2021-02-03 | 2021-06-15 | 上海明略人工智能(集团)有限公司 | Sales lead information identification method and system based on voice identification technology |
| CN112967721B (en) * | 2021-02-03 | 2024-05-31 | 上海明略人工智能(集团)有限公司 | Sales lead information recognition method and system based on voice recognition technology |
| CN112861005A (en) * | 2021-02-22 | 2021-05-28 | 上海风秩科技有限公司 | Method, device and equipment for information push |
| CN112950326A (en) * | 2021-03-17 | 2021-06-11 | 南通大学 | Artificial intelligence data analysis system supporting deep learning working principle |
| CN113282783A (en) * | 2021-05-28 | 2021-08-20 | 上海明略人工智能(集团)有限公司 | Method and device for information recommendation, electronic equipment and readable storage medium |
| CN113538048A (en) * | 2021-07-12 | 2021-10-22 | 深圳市明源云客电子商务有限公司 | Demand information obtaining method and device, terminal equipment and storage medium |
| CN113593552A (en) * | 2021-07-12 | 2021-11-02 | 深圳市明源云客电子商务有限公司 | Demand information statistical method and device, terminal equipment and storage medium |
| CN113836379B (en) * | 2021-09-26 | 2023-08-25 | 北京百炼智能科技有限公司 | Intelligent recommendation method and system based on client image |
| CN113836379A (en) * | 2021-09-26 | 2021-12-24 | 北京百炼智能科技有限公司 | Intelligent recommendation method and system based on customer image |
| CN114866818A (en) * | 2022-06-17 | 2022-08-05 | 深圳壹账通智能科技有限公司 | Video recommendation method and device, computer equipment and storage medium |
| CN114866818B (en) * | 2022-06-17 | 2024-04-26 | 深圳壹账通智能科技有限公司 | Video recommendation method, device, computer equipment and storage medium |
| CN116303975B (en) * | 2023-05-11 | 2023-07-21 | 腾讯科技（深圳）有限公司 | Training method of recall model, recall method and related equipment |
| CN116303975A (en) * | 2023-05-11 | 2023-06-23 | 腾讯科技（深圳）有限公司 | Training method of recall model, recall method and related equipment |
| CN117312878A (en) * | 2023-10-30 | 2023-12-29 | 青岛汇聚融合健康科技有限公司 | Method, system and device for generating mall user mark |
| CN117312878B (en) * | 2023-10-30 | 2024-05-14 | 青岛汇聚融合健康科技有限公司 | Method, system and device for generating mall user mark |
| CN117710122A (en) * | 2024-01-11 | 2024-03-15 | 广州小锤科技服务有限公司 | Big data claim product management method, device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109727092A (en) | Products Show method, apparatus, computer equipment and storage medium based on AI | |
| CN109147934B (en) | Inquiry data recommendation method, device, computer equipment and storage medium | |
| US10733197B2 (en) | Method and apparatus for providing information based on artificial intelligence | |
| US20180225363A1 (en) | System and Methods for Automating Trademark and Service Mark Searches | |
| CN110674271B (en) | Question and answer processing method and device | |
| CN109492164A (en) | A kind of recommended method of resume, device, electronic equipment and storage medium | |
| CN112711937A (en) | Template recommendation method, device, equipment and storage medium | |
| CN102663129A (en) | Medical field deep question and answer method and medical retrieval system | |
| CN109800307A (en) | Analysis method, device, computer equipment and the storage medium of product evaluation | |
| CN109960756A (en) | Media event information inductive method | |
| CN102955848A (en) | Semantic-based three-dimensional model retrieval system and method | |
| CN109325115B (en) | Role analysis method and analysis system | |
| CN106951503A (en) | Information providing method, device, equipment and storage medium | |
| CN111339284A (en) | Product intelligent matching method, device, equipment and readable storage medium | |
| CN104077327B (en) | The recognition methods of core word importance and equipment and search result ordering method and equipment | |
| CN113409907A (en) | Intelligent pre-inquiry method and system based on Internet hospital | |
| Mohapatra et al. | Using TF-IDF on Kisan call centre dataset for obtaining query answers | |
| US20130332440A1 (en) | Refinements in Document Analysis | |
| CN109471927A (en) | A kind of knowledge base and its foundation, answering method and application apparatus | |
| Ozmutlu et al. | Are people asking questions of general Web search engines? | |
| US20190080250A1 (en) | Identification of a Series of Compatible Components Using Artificial Intelligence | |
| US11694033B2 (en) | Transparent iterative multi-concept semantic search | |
| WO2019053488A1 (en) | Identification of a series of compatible components using artificial intelligence | |
| CN112632246A (en) | Robot dialogue method and device based on deep learning and computer equipment | |
| CN113971581A (en) | Robot control method and device, terminal equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |