WO2020253363A1 - 产品认可度分析方法、装置、终端及计算机可读存储介质 - Google Patents

产品认可度分析方法、装置、终端及计算机可读存储介质 Download PDF

Info

Publication number
WO2020253363A1
WO2020253363A1 PCT/CN2020/086166 CN2020086166W WO2020253363A1 WO 2020253363 A1 WO2020253363 A1 WO 2020253363A1 CN 2020086166 W CN2020086166 W CN 2020086166W WO 2020253363 A1 WO2020253363 A1 WO 2020253363A1
Authority
WO
WIPO (PCT)
Prior art keywords
product
micro
recognition
expression
instant
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.)
Ceased
Application number
PCT/CN2020/086166
Other languages
English (en)
French (fr)
Inventor
张美苑
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
OneConnect Smart Technology Co Ltd
Original Assignee
OneConnect Smart Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by OneConnect Smart Technology Co Ltd filed Critical OneConnect Smart Technology Co Ltd
Publication of WO2020253363A1 publication Critical patent/WO2020253363A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G06Q30/0282—Rating or review of business operators or products
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/06—Asset management; Financial planning or analysis
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/174—Facial expression recognition

Definitions

  • This application relates to the field of artificial intelligence technology, and in particular to a product recognition analysis method, device, terminal, and computer-readable storage medium.
  • the business personnel of financial companies at this stage need to know whether the recommendation of the financial product is successful after introducing the financial product to the customer, that is, they need to know whether the customer approves the financial product introduced.
  • the existing way to obtain product approval from customers is to conduct manual return visits to customers.
  • the inventor realizes that the manual research method requires a lot of labor costs, which makes the analysis of product recognition inefficient.
  • a method, device, terminal, and computer-readable storage medium for analyzing product recognition are provided.
  • a product recognition analysis method includes:
  • the customer's recognition of the product is obtained.
  • This application also provides a product recognition analysis device, the product recognition analysis device includes:
  • a first acquisition module the first acquisition module is used to acquire video images in a product introduction time period
  • a recognition module configured to recognize the video image to obtain the micro expression type corresponding to the facial micro expression in the video image and the instantaneous confidence level corresponding to the micro expression type;
  • a second acquisition module configured to acquire a corresponding correlation coefficient according to the type of the micro expression
  • a calculation module which is used to fill the instantaneous confidence and the correlation coefficient into the investment willingness formula to obtain the instant recognition of the product
  • the third obtaining module is used to obtain the customer's recognition of the product according to the instant recognition of the product within the product introduction time period.
  • the present application also provides a terminal, including a processor, a memory, and a product approval analysis program stored on the memory that can be executed by the processor, wherein the product approval analysis program is executed by the processor When executing, implement the steps of the following method:
  • the customer's recognition of the product is obtained.
  • the present application also provides a computer-readable storage medium having a product recognition analysis program stored on the computer-readable storage medium, wherein the product recognition analysis program is executed by a processor to implement the steps of the following method.
  • the customer's recognition of the product is obtained.
  • FIG. 1 is a schematic diagram of the hardware structure of a terminal involved in a solution of an embodiment of the application
  • FIG. 2 is a schematic flowchart of the first embodiment of the product recognition analysis method of the application
  • FIG. 3 is a detailed flow diagram of the steps of recognizing the video image to obtain the micro-expression type corresponding to the facial micro-expression in the video image and the instantaneous confidence level corresponding to the micro-expression type in the embodiment of the application;
  • FIG. 4 is a detailed schematic diagram of the process of obtaining the customer's degree of recognition of the product according to the instantaneous recognition of the product within the product introduction time period in an embodiment of the application;
  • FIG. 5 is a schematic flowchart of a second embodiment of the product recognition analysis method of the application.
  • FIG. 6 is a schematic flowchart of a third embodiment of the product recognition analysis method of the application.
  • FIG. 7 is a schematic flowchart of a fourth embodiment of the product recognition analysis method of this application.
  • Figure 8 is a schematic diagram of modules of the product recognition analysis device of the application.
  • the product recognition analysis method involved in the embodiment of the present application is mainly applied to a terminal, and the terminal may be a device with display and processing functions such as a PC, a portable computer, and a mobile terminal.
  • FIG. 1 is a schematic diagram of a terminal structure involved in a solution of an embodiment of this application.
  • the terminal may include a processor 1001 (for example, a CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005.
  • the communication bus 1002 is used to realize the connection and communication between these components;
  • the user interface 1003 may include a display (Display), an input unit such as a keyboard (Keyboard);
  • the network interface 1004 may optionally include a standard wired interface, a wireless interface (Such as WI-FI interface);
  • the memory 1005 can be a high-speed RAM memory or a non-volatile memory, such as a disk memory.
  • the memory 1005 can optionally be a storage device independent of the aforementioned processor 1001 .
  • FIG. 1 does not constitute a limitation on the device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange different components.
  • the memory 1005 as a computer-readable storage medium in FIG. 1 may include an operating system, a network communication module, and a product recognition analysis program.
  • the network communication module is mainly used to connect to the server and perform data communication with the server; and the processor 1001 can call the product recognition analysis program stored in the memory 1005 and execute the steps of the product recognition analysis method.
  • This application provides a product recognition analysis method.
  • the product recognition analysis method includes the following steps:
  • Step S100 obtaining video images during the product introduction time period
  • the customer's facial expression can be recorded.
  • the video image during the product introduction period can be obtained.
  • the product recognition micro-expression model in the terminal, and the video image during the product introduction time period is input into the product recognition micro-expression model, and the terminal will obtain the video image during the product introduction time period to provide information on the product introduction time period Analysis of video images.
  • Step S200 Recognizing the video image to obtain the micro-expression type corresponding to the facial micro-expression in the video image and the instantaneous confidence level corresponding to the micro-expression type;
  • the product recognition micro-expression model can recognize video images and can obtain 54 types of micro-expression.
  • the micro-expression types include love, interest, surprise, expectation...aggressiveness, conflict 54 types of micro expressions including, insult, doubt and fear.
  • the instantaneous confidence level corresponding to the micro-expression type refers to the probability that the facial micro-expression belongs to a certain micro-expression type, and the micro-expression confidence rate returned by the interface is returned in real time, so it is called the instantaneous confidence rate.
  • the probability of recognizing that the video image belongs to the micro-expression type "love” is 0.9, and the instantaneous confidence of the micro-expression type belonging to "love” is 0.9, and it belongs to "love”.
  • the probability of the micro-expression type of "suspicion” is 0.05, and the instantaneous confidence of "suspicion” is 0.05.
  • the product recognition micro-expression model will recognize the video image, and the micro-expression type and micro-expression corresponding to the facial micro-expression of each frame of the video image The instantaneous confidence level corresponding to the type.
  • step S200 includes:
  • Step S210 Recognizing the video image to obtain the micro-expression feature corresponding to the facial micro-expression in the video image;
  • each micro-expression type corresponds to a specific micro-expression feature. Therefore, after the video image is obtained, the video image is recognized, and the micro-expression corresponding to the facial micro-expression in each frame of the video image is obtained feature.
  • Step S220 Obtain the corresponding micro expression type and the instantaneous confidence level corresponding to the micro expression type according to the micro expression feature.
  • the micro-expression type corresponding to the micro-expression feature can be obtained, and the probability that the facial micro-expression belongs to the micro-expression type can be identified. It should be noted that during the recognition process, if there is no specific micro-expression feature in the video image, the probability (ie, instantaneous confidence) of identifying the type of micro-expression corresponding to the specific micro-expression feature is 0, that is, No need to consider in the subsequent analysis process.
  • Step S300 Obtain a corresponding correlation coefficient according to the micro-expression type
  • a micro-expression information table is preset in the terminal. As shown in Table I, after obtaining the micro-expression type corresponding to the video image, the corresponding correlation coefficient can be found in the micro-expression information table.
  • the correlation coefficient refers to The degree of correlation between the type of micro-expression corresponding to the video image and the willingness of potential customers to invest in the product.
  • the correlation coefficient is specifically Pearson Correlation Coefficient (Pearson Correlation Coefficient) is used to measure whether two data sets are on a line, and it is used to measure the linear relationship between the distance variables. In this embodiment, the value of the correlation coefficient is [-1, 1].
  • correlation coefficient 0.8-1.0 indicates very strong correlation
  • correlation coefficient 0.6-0.8 indicates strong correlation
  • correlation coefficient 0.4-0.6 indicates moderate correlation
  • correlation coefficient 0.2-0.4 indicates weak Correlation
  • correlation coefficient 0.0-0.2 indicates very weak correlation or no correlation.
  • Table I Micro expression information table
  • positive Category refers to items that have a positive effect on investment willingness and are strongly correlated: such as (love), (happy joy), (optimism), (trust), (acceptance), (surprise), etc.
  • negative category refers to items that have a negative effect on investment willingness and are strongly correlated: such as (distraction), (defiance against Defiance), (apprehension), (sadness), (annoyance); weakly related categories refer to A neutral correlation item that has a general effect on investment willingness.
  • Step S400 filling the instantaneous confidence and the correlation coefficient into the investment willingness degree formula to obtain the instant recognition degree of the product
  • the instantaneous confidence and the correlation coefficient are filled into the investment willingness formula to obtain the instant recognition of the product.
  • the product recognition micro expression model recognizes that the product recognition micro expression value of the face image in each frame of video image is a score with a value range of [-1, 1], the micro expression confidence rate returned by the interface is It is returned in real time, so this score is also an instant value of the user's recognition of the product, which is called the instant recognition of the product.
  • p i refers to the correlation coefficient corresponding to the micro-expression type i
  • q i refers to the instantaneous instantaneous confidence level corresponding to the micro-expression type i
  • R refers to the instant recognition degree of the product.
  • the product recognition level can be preset in the terminal.
  • the corresponding product recognition level is very high; when the user’s product recognition level is [0.3, 0.8]
  • the corresponding product recognition degree is lower; when the user’s product recognition degree is [-1, -0.8], the corresponding product recognition level is very low.
  • step S500 the customer's degree of recognition of the product is obtained according to the instantaneous recognition of the product within the product introduction time period.
  • the customer's recognition of the product can be obtained according to all the instant recognition of the products.
  • step S500 includes :
  • Step S510 obtaining the product instant recognition degree of a single customer for the product during the product introduction time period
  • the instantaneous degree of product recognition of a single customer for the product during the product introduction period can be obtained.
  • Step S520 drawing a micro-expression trend chart of a single customer on the product during the product introduction period according to the instant recognition of the product;
  • micro-expression trend chart of a single customer for the product during the product introduction period.
  • the abscissa of the micro-expression trend chart represents the time, and the ordinate represents the instant recognition of the product.
  • the trend chart can clearly see the customer's micro-expression changes during the introduction of the product by the business staff.
  • Step S530 Obtain the approval level of a single customer for the product according to the micro-expression trend chart.
  • the customer's recognition level of the product can be obtained according to the trend.
  • the user’s product recognition is mostly concentrated in [0.8, 1] during the product introduction time period, the user’s product recognition for the product is very high; when the product introduction time period, the user’s product recognition instantly
  • the degree of recognition is mostly concentrated in [0.3, 0.8]
  • the user’s product recognition for the product is relatively high; during the product introduction period, the user’s product instant recognition is mostly concentrated in [-0.8, 0.3]
  • the user’s product recognition for the product is relatively low; when the user’s product recognition is mostly concentrated in [-1, -0.8] during the product introduction period, the user’s product recognition for the product The recognition level is very low.
  • video images during the product introduction time period are acquired; the video images are recognized to acquire the micro-expression type corresponding to the facial micro-expression in the video image and the instantaneous confidence corresponding to the micro-expression type; and the corresponding micro-expression type is obtained
  • the correlation coefficient of the product fill the instantaneous confidence and correlation coefficient into the investment willingness formula to obtain the instant recognition of the product; obtain the customer’s recognition of the product according to the instant recognition of the product during the product introduction period.
  • the technical solution proposed in this application is based on micro-expression recognition to recognize facial micro-expressions in video images.
  • the instantaneous confidence and correlation coefficients corresponding to the micro-expression types are filled into the investment willingness formula to obtain the instant recognition of the product.
  • the instant recognition degree obtains the degree of customer recognition of the product, which can better and more accurately guide the financial business personnel to rationally allocate marketing activities.
  • the recognition of the product by the customer can be obtained by recognizing the video image, and the analysis efficiency of the recognition of the product can be improved.
  • FIG. 5 is a schematic flowchart of the second embodiment of the product recognition analysis method of this application. Based on the above embodiment, after step S520, the method further includes:
  • Step S521 Obtain sales vocabulary during the product introduction time period
  • the business personnel introduce the various situations of the product to the customer, among which various sales techniques (topics) are designed.
  • various sales techniques topics
  • Step S522 Obtain the micro expression trend of each product under the sales language according to the micro expression trend graph of the product;
  • the corresponding time period is found in the micro expression trend chart, and the customer's micro expression trend in the corresponding time period is obtained, that is, obtain The micro-expression trend of the product under each sales language, so as to help business personnel quickly grasp the change trend of the customer's product recognition under the sales language.
  • Step S523 Obtain key words according to the micro-expression trend of the products under each sales word, and add the key words to the smart words library.
  • the corresponding sales word in the time period when the micro-expression trend rises is used as the key word, that is, the corresponding sales word in the time period when the customer’s product recognition rises instantly Sales words are used as key words, and then the key words are added to the smart words library.
  • Adding key words that can improve customer recognition to the smart phone library allows business personnel to adjust sales skills and service quality in time according to the key words in the smart phone library to achieve sales intentions. That is, through the intelligent speech library to feed back the business staff's own speech skills, so as to improve the sales level of the business staff, save the cost of speech training for the business staff, and remove the large amount of manpower and material costs for customer return visits.
  • FIG. 6 is a schematic flowchart of the third embodiment of the product recognition analysis method of this application. Based on the above embodiment, after step S530, it further includes:
  • Step S540 obtaining the recognition level of all customers for the same product during the product introduction time period
  • the product recognition level is divided into very high, high, low, and very low.
  • the terminal After obtaining the recognition level of a single customer for the product through the micro-expression trend chart, the terminal can obtain all customers' recognition of the same product during the product introduction period.
  • the recognition level of the product for example, when the introduced products are A, B, C, all the customers who have been promoted by the business personnel have the recognition level of A, the recognition level of B and the recognition of C Degree level.
  • Step S550 Count the number of customers with different product recognition levels
  • Step S560 Obtain the degree of market recognition of the same product according to the number of customers.
  • the recognition level of the product in the entire market can be judged according to the number of customers under each product recognition level. If the product recognition level is a very high number of customers At most, it means that the entire market has a high degree of recognition of the product; if the product recognition level is the highest number of customers, it means that the entire market has a higher recognition level of the product; if the product recognition level is the lower number of customers, It means that the whole market has a low degree of product recognition; if the number of customers with a very low product recognition level is the largest, it means that the whole market has a low degree of product recognition.
  • FIG. 7 is a schematic flowchart of the fourth embodiment of the product recognition analysis method of this application. Based on the first embodiment, after step S530, it further includes:
  • Step S570 obtaining the recognition level of a single customer for different products during the product introduction time period
  • the video images when the business personnel introduce products to customers can be recognized to obtain the instant recognition of the customers' products. Therefore, the video images of different products introduced by the same customer can be identified, and the video images of different products introduced by the same customer After all the images are recognized, the recognition level of a single customer for different products can be obtained.
  • Step S580 drawing a distribution map of a single customer's preference for different products according to the recognition level of a single customer for different products;
  • Step S590 Acquire the customer's product preference according to the preference distribution map.
  • the product recognition levels of all customers for different products can also be obtained, and all customers are distinguished according to age, so as to obtain the product recognition levels of customers of different ages for different products.
  • product A obtain the product recognition levels of customers of different ages for product A, and calculate the age distribution of customers under different product recognition levels, and which age group of customers have high product recognition levels for A product The proportion of customers is the largest, indicating that A product is more popular in this age group.
  • the micro-expression image samples when building the product recognition degree micro-expression model, can be collected first, and the micro-expression image samples are input into the product recognition micro-expression model through the Gamma micro-expression interface, and the original micro-expression recognition
  • the model performs feature extraction, clustering and setting correlation coefficients on the micro-expression image samples to generate the investment willingness formula.
  • the product recognition micro-expression model of customer recognition of financial products is trained through convolutional neural networks, and the formula is continuously corrected to continuously improve the accuracy.
  • the correction process of the model formula is based on the product recognition micro-expression value data of users who actually purchased the product (indicating that the product recognition is high) and the product recognition micro-expression value of customers who have not purchased the product (indicating that the product recognition is very low) Data and facial expression pictures, obtain the instantaneous confidence rate corresponding to the actual micro-expression type returned, and then adjust the correlation coefficient in the formula to optimize the feedback to improve the accuracy of the model.
  • this application also provides a product approval degree analysis device 10, the product approval degree analysis device 10 includes:
  • a first acquisition module, 20, the first acquisition module is used to acquire video images within a product introduction time period;
  • the recognition module 30 is configured to recognize the video image to obtain the micro expression type corresponding to the facial micro expression in the video image and the instantaneous confidence level corresponding to the micro expression type;
  • the second obtaining module 40 is configured to obtain the corresponding correlation coefficient according to the type of the micro expression
  • a calculation module 50 which is used to fill the instantaneous confidence and the correlation coefficient into the investment willingness formula to obtain the instant recognition of the product
  • the third obtaining module 60 is used to obtain the customer's recognition of the product according to the instant recognition of the product within the product introduction time period.
  • the third acquiring module 60 is further used for:
  • the product recognition analysis device 10 further includes:
  • a fourth acquisition module is used to acquire sales words within a product introduction time period
  • the fifth obtaining module is used to obtain the micro expression trend of each product under the sales language according to the micro expression trend graph of the product;
  • the adding module is used to obtain key words according to the micro expression trend of the products under each sales word, and add the key words to the smart words library.
  • the product recognition analysis device 10 further includes:
  • a sixth obtaining module is used to obtain the recognition level of all customers for the same product during the product introduction time period;
  • a statistics module which is used to count the number of customers with different product recognition levels
  • the seventh obtaining module is configured to obtain the degree of market recognition for the same product according to the number of customers.
  • the product recognition analysis device 10 further includes:
  • An eighth obtaining module is used to obtain the recognition level of a single customer for different products during a product introduction time period;
  • a drawing module which is used to draw a single customer’s preference distribution map for different products according to the level of recognition of different products by a single customer;
  • the ninth obtaining module is configured to obtain the customer's product preference according to the preference distribution map.
  • the identification module 30 is further used for:
  • each module in the aforementioned product recognition analysis device 10 corresponds to each step in the aforementioned embodiment of the product recognition analysis method, and its functions and implementation processes are not repeated here.
  • this application also provides a computer-readable storage medium.
  • the computer-readable storage medium may be a non-volatile storage medium or a volatile storage medium.
  • the computer-readable storage medium of the present application stores a product recognition degree analysis program, where the product recognition degree analysis program is executed by a processor to realize the steps of the product recognition degree analysis method described above.
  • this application can be provided as methods, systems, or computer program products. Therefore, this application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, this application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.
  • computer-usable storage media including but not limited to disk storage, CD-ROM, optical storage, etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Finance (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • General Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Economics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Multimedia (AREA)
  • Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Operations Research (AREA)
  • Technology Law (AREA)
  • General Engineering & Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Human Computer Interaction (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种产品认可度分析方法、装置、终端及计算机可读存储介质。方法包括：获取产品介绍时间段内的视频图像（S100）；对视频图像进行识别获取视频图像中人脸微表情对应的微表情类型以及微表情类型对应的瞬时置信度（S200）；根据微表情类型获取对应的相关系数（S300）；将瞬时置信度和相关系数填入投资意愿度公式，得到产品瞬间认可度（S400）；根据产品介绍时间段内的产品瞬间认可度获得客户对产品的认可程度（S500）。该方法基于微表情识别对视频图像中人脸微表情进行识别，来获得客户对产品的认可程度，通过对视频图像进行识别就能获得客户对产品的认可度，能够提高对产品的认可度的分析效率。

Description

产品认可度分析方法、装置、终端及计算机可读存储介质
本申请要求于2019年6月19日提交中国专利局，申请号为201910532837.3、发明名称为“产品认可度分析方法、装置、终端及计算机可读存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及人工智能技术领域，尤其涉及一种产品认可度分析方法、装置、终端及计算机可读存储介质。
背景技术
目前，现阶段的金融企业的业务人员在给客户介绍完金融产品后，需要知道金融产品的推荐是否成功，即，需要知道客户对介绍的金融产品是否认可。现有的获得客户对产品是否认可的方式是对客户进行人工回访调研，但是，发明人意识到，人工调研的方式需要耗费大量的人力成本，使得产品认可度分析效率低。
因此，现有的产品认可度分析效率低是一种亟待解决的问题。
发明内容
根据本申请公开的各种实施例，提供一种产品认可度分析方法、装置、终端及计算机可读存储介质。
一种产品认可度分析方法，所述产品认可度分析方法包括：
获取产品介绍时间段内的视频图像；
对所述视频图像进行识别获取所述视频图像中人脸微表情对应的微表情类型以及所述微表情类型对应的瞬时置信度；
根据所述微表情类型获取对应的相关系数；
将所述瞬时置信度和所述相关系数填入投资意愿度公式，得到产品瞬间认可度；
根据产品介绍时间段内的产品瞬间认可度获得客户对产品的认可程度。
本申请还提供一种产品认可度分析装置，所述产品认可度分析装置包括：
第一获取模块，所述第一获取模块用于获取产品介绍时间段内的视频图像；
识别模块，所述识别模块用于对所述视频图像进行识别获取所述视频图像中人脸微表情对应的微表情类型以及所述微表情类型对应的瞬时置信度；
第二获取模块，所述第二获取模块用于根据所述微表情类型获取对应的相关系数；
计算模块，所述计算模块用于将所述瞬时置信度和所述相关系数填入投资意愿度公式，得到产品瞬间认可度；
第三获取模块，所述第三获取模块用于根据产品介绍时间段内的产品瞬 间认可度获得客户对产品的认可程度。
本申请还提供一种终端，包括处理器、存储器、以及存储在所述存储器上的可被所述处理器执行的产品认可度分析程序，其中，所述产品认可度分析程序被所述处理器执行时，实现如下方法的步骤：
获取产品介绍时间段内的视频图像；
对所述视频图像进行识别获取所述视频图像中人脸微表情对应的微表情类型以及所述微表情类型对应的瞬时置信度；
根据所述微表情类型获取对应的相关系数；
将所述瞬时置信度和所述相关系数填入投资意愿度公式，得到产品瞬间认可度；
根据产品介绍时间段内的产品瞬间认可度获得客户对产品的认可程度。
本申请还提供一种计算机可读存储介质，所述计算机可读存储介质上存储有产品认可度分析程序，其中，所述产品认可度分析程序被处理器执行时，实现如下方法的步骤。
获取产品介绍时间段内的视频图像；
对所述视频图像进行识别获取所述视频图像中人脸微表情对应的微表情类型以及所述微表情类型对应的瞬时置信度；
根据所述微表情类型获取对应的相关系数；
将所述瞬时置信度和所述相关系数填入投资意愿度公式，得到产品瞬间认可度；
根据产品介绍时间段内的产品瞬间认可度获得客户对产品的认可程度。
本申请的一个或多个实施例的细节在下面的附图和描述中提出。本申请的其它特征和优点将从说明书、附图以及权利要求书变得明显。
附图说明
图1为本申请实施例方案中涉及的终端的硬件结构示意图；
图2为本申请产品认可度分析方法第一实施例的流程示意图；
图3为本申请实施例中对所述视频图像进行识别获取所述视频图像中人脸微表情对应的微表情类型以及所述微表情类型对应的瞬时置信度的步骤的流程细化示意图；
图4为本申请实施例中根据产品介绍时间段内的产品瞬间认可度获得客户对产品的认可程度的步骤的流程细化示意图；
图5为本申请产品认可度分析方法第二实施例的流程示意图；
图6为本申请产品认可度分析方法第三实施例的流程示意图；
图7为本申请产品认可度分析方法第四实施例的流程示意图；
图8为本申请产品认可度分析装置的模块示意图。
本申请目的的实现、功能特点及优点将结合实施例，参照附图做进一步说明。
具体实施方式
应当理解，此处所描述的具体实施例仅仅用以解释本申请，并不用于限 定本申请。
本申请实施例涉及的产品认可度分析方法主要应用于终端，该终端可以是PC、便携计算机、移动终端等具有显示和处理功能的设备。
参照图1，图1为本申请实施例方案中涉及的终端结构示意图。本申请实施例中，终端可以包括处理器1001(例如CPU)，通信总线1002，用户接口1003，网络接口1004，存储器1005。其中，通信总线1002用于实现这些组件之间的连接通信；用户接口1003可以包括显示屏(Display)、输入单元比如键盘(Keyboard)；网络接口1004可选的可以包括标准的有线接口、无线接口(如WI-FI接口)；存储器1005可以是高速RAM存储器，也可以是稳定的存储器(non-volatile memory)，例如磁盘存储器，存储器1005可选的还可以是独立于前述处理器1001的存储装置。
本领域技术人员可以理解，图1中示出的硬件结构并不构成对设备的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。
继续参照图1，图1中作为一种计算机可读存储介质的存储器1005可以包括操作系统、网络通信模块以及产品认可度分析程序。
在图1中，网络通信模块主要用于连接服务器，与服务器进行数据通信；而处理器1001可以调用存储器1005中存储的产品认可度分析程序，并执行产品认可度分析方法的步骤。
基于上述终端的硬件结构，提出本申请产品认可度分析方法的各个实施例。
本申请提供一种产品认可度分析方法。
请参阅图2，在本申请第一实施例中，产品认可度分析方法包括以下步骤：
步骤S100，获取产品介绍时间段内的视频图像；
具体地，可以在金融行业的业务人员给客户介绍金融产品时，对客户的面部表情进行录像，在业务人员介绍完毕后，可以得到产品介绍时间段内的视频图像。在终端中设有产品认可度微表情模型，将产品介绍时间段内的视频图像输入产品认可度微表情模型中，终端将获取到产品介绍时间段内的视频图像，以对产品介绍时间段内的视频图像进行分析。
步骤S200，对所述视频图像进行识别获取所述视频图像中人脸微表情对应的微表情类型以及所述微表情类型对应的瞬时置信度；
需要说明的是，产品认可度微表情模型可以对视频图像进行识别，可以获得54种微表情类型，如表I所示，微表情类型包括爱、感兴趣、惊喜、期待……攻击性、冲突、侮辱、怀疑和恐惧等54种微表情类型。微表情类型对应的瞬时置信度是指人脸微表情属于某一种微表情类型的概率，而接口返回来的微表情置信率是实时返回的，所以称为瞬时置信率。例如，对一帧视频图像进行识别时，识别到该视频图像属于“爱”这一微表情类型的概率为0.9，则属于“爱”这一微表情类型的瞬时置信度为0.9，而属于“怀疑”这一微表情类型的概率为0.05，则属于“怀疑”的瞬时置信度为0.05。
具体地，在获取到产品介绍时间段内的视频图像后，产品认可度微表情模型将对视频图像进行识别，视频图像中的每一帧图像的人脸微表情对应的微表情类型以及微表情类型对应的瞬时置信度。
具体地，请参照图3，图3为本申请实施例中对所述视频图像进行识别获取所述视频图像中人脸微表情对应的微表情类型以及所述微表情类型对应的瞬时置信度的步骤的流程细化示意图，基于上述实施例，步骤S200包括：
步骤S210，对所述视频图像进行识别获取所述视频图像中人脸微表情对应的微表情特征；
可以理解的是，每一种微表情类型都对应有特定的微表情特征，因此，在获取到视频图像后，对视频图像进行识别，获取每一帧视频图像中人脸微表情对应的微表情特征。
步骤S220，根据所述微表情特征获得对应的微表情类型以及所述微表情类型对应的瞬时置信度。
在获得微表情特征后，可以得到该微表情特征对应的微表情类型，同时识别出该人脸微表情属于该微表情类型的概率。需要说明的是，在识别过程中，若视频图像中不存在特定的微表情特征，则识别出与该特定的微表情特征对应的微表情类型的概率(即瞬时置信度)为0，即，在后续分析过程中无需考虑。
步骤S300，根据所述微表情类型获取对应的相关系数；
具体地，在终端中预先设有微表情信息表，如表I所示，在获取到视频图像对应的微表情类型后，可以在微表情信息表中查找到对应的相关系数，相关系数是指视频图像对应的微表情类型与潜在客户对产品的投资意愿的相关程度。相关系数具体为Pearson相关系数(Pearson CorrelationCoefficient)是用来衡量两个数据集合是否在一条线上面，它用来衡量定距变量间的线性关系。本实施例中，相关系数的取值在【-1，1】，相关系数的绝对值越大，相关性越强；即，相关系数越接近于1或-1，相关度越强；相关系数越接近于0，相关度越弱。通常情况下通过以下取值范围判断变量的相关强度：相关系数0.8-1.0表示极强相关；相关系数0.6-0.8表示强相关；相关系数0.4-0.6表示中等程度相关；相关系数0.2-0.4表示弱相关；相关系数0.0-0.2表示极弱相关或无相关。
表I：微表情信息表
Figure PCTCN2020086166-appb-000001
Figure PCTCN2020086166-appb-000002
Figure PCTCN2020086166-appb-000003
其中，该微表情信息表中，将54种微表情划分为积极类别、消极类别、介于积极类别和消极类别之间的弱相关类别(即微表情类型为性格描述词及中性词，积极类别是指对投资意愿度起积极作用且强关联项：如(喜爱Love),(开心joy),(乐观optimism),(信任tust)、(可接受acceptance),(惊喜surprise)等；消极类别是指对投资意愿度起消极作用且强关联项：如(心不在焉distraction)、(抵触违抗Defiance)、(愁眉不展Apprehension)、(悲伤sadness)、(烦心倦目Annoyance)；弱相关类别是指对投资意愿度起一般作用的中性关联项。
步骤S400，将所述瞬时置信度和所述相关系数填入投资意愿度公式，得到产品瞬间认可度；
具体地，在获取到微表情类型对应的瞬时置信率和微表情类型对应相关系数后，将瞬时置信度和相关系数填入投资意愿度公式，得到产品的产品瞬间认可度。由于产品认可度微表情模型识别到每一帧视频图像中人脸图像的产品认可度微表情值是一个取值范围为【-1，1】的得分，而接口返回来的微表情置信率是实时返回的，所以这个得分也是用户对该产品的认可度的一个瞬间值，称为产品瞬间认可度。
其中，投资意愿度公式为
Figure PCTCN2020086166-appb-000004
其中，p i是指微表情类型i对应的相关系数，q i是指微表情类型i对应的瞬间瞬时置信度，R是指产品瞬间认可度。
例如，当对一帧视频图像进行识别时，识别到了“爱”与“怀疑”这两种微表情类型，并且，识别到该视频图像属于“爱”这一微表情类型的瞬时置信度为0.9，而属于“怀疑”这一微表情类型的瞬时置信度为0.05，通过查询，“爱”这一微表情类型对应的相关系数为1，“怀疑”这一微表情类型对应的相关系数为-0.5，那么，在该视频图像下，客户对产品的产品瞬间认可度为：R＝0.9*1+0.05*(-0.5)＝0.875。
此外，可以在终端预设产品认可度等级，当用户的产品瞬间认可度为【0.8，1】时，对应的产品认可度等级为很高；当用户的产品瞬间认可度为【0.3，0.8】时，对应的产品认可度等级为较高；当用户的产品瞬间认可度为【-0.8，0.3】时，对应的产品认可度等级为较低；当用户的产品瞬间认可度为【-1，-0.8】时，对应的产品认可度等级为很低。
步骤S500，根据产品介绍时间段内的产品瞬间认可度获得客户对产品的认可程度。
在获得产品介绍时间段内的所有的产品瞬间认可度后，可以根据所有的产品瞬间认可度可以获得客户对产品的认可程度。
具体地，请参照图4，图4为本申请实施例中根据产品介绍时间段内的产 品瞬间认可度获得客户对产品的认可程度的步骤的流程细化示意图，基于上述实施例，步骤S500包括：
步骤S510，获取产品介绍时间段内单个客户对产品的产品瞬间认可度；
具体地，为了获得单个客户对单个产品的认可程度，可以获取产品介绍时间段内单个客户对产品的产品瞬间认可度。
步骤S520，根据所述产品瞬间认可度绘制出产品介绍时间段内单个客户对产品的微表情走势图；
具体地，根据获取到的产品的瞬间认可度绘制出产品介绍时间段内单个客户对产品的微表情走势图，微表情走势图的横坐标表示时间，纵坐标表示产品瞬间认可度，通过微表情走势图可以清晰的看到客户在业务人员介绍产品期间的微表情变化情况。
步骤S530，根据所述微表情走势图获取单个客户对产品的认可程度等级。
具体地，在绘制出微表情走势图后，根据走势可以得出客户对产品的认可程度等级。当产品介绍时间段内，用户的产品瞬间认可度大部分集中分布在【0.8，1】时，则该用户对该产品的产品认可程度很高；当产品介绍时间段内，用户的产品瞬间认可度大部分集中分布在【0.3，0.8】时，则该用户对该产品的产品认可程度较高；当产品介绍时间段内，用户的产品瞬间认可度大部分集中分布在【-0.8，0.3】时，则该用户对该产品的产品认可程度较低；当产品介绍时间段内，用户的产品瞬间认可度大部分集中分布在【-1，-0.8】时，则该用户对该产品的产品认可程度很低。
本申请技术方案中，获取产品介绍时间段内的视频图像；对视频图像进行识别获取视频图像中人脸微表情对应的微表情类型以及微表情类型对应的瞬时置信度；根据微表情类型获取对应的相关系数；将瞬时置信度和相关系数填入投资意愿度公式，得到产品瞬间认可度；根据产品介绍时间段内的产品瞬间认可度获得客户对产品的认可程度。本申请提出的技术方案基于微表情识别对视频图像中人脸微表情进行识别，将与微表情类型对应的瞬时置信度和相关系数填入投资意愿度公式，得到产品瞬间认可度，并根据产品瞬间认可度获得客户对产品的认可程度，能够更好更精准地指导金融业务人员合理配置营销活动。本申请通过对视频图像进行识别就能获得客户对产品的认可度，能够提高对产品的认可度的分析效率。
在其中一个实施例中，请参照图5，图5为本申请产品认可度分析方法第二实施例的流程示意图，基于上述实施例，步骤S520之后，还包括：
步骤S521，获取产品介绍时间段内的销售话术；
具体地，在产品介绍时间段内业务人员向客户介绍产品的各种情况，其中，设计到各种销售话术(话题)，例如，在介绍产品A时，需要有产品特征、收益普及、风险普及、产品特色、意向询问、拓展介绍、结束语等销售话术。在绘制出客户的微表情走势图后，获取产品介绍时间段内的销售话术。
步骤S522，根据产品的微表情走势图获得每个销售话术下产品的微表情走势；
具体地，获取到销售话术后，根据业务人员说出销售话术的时间段，在微表情走势图中找到对应的时间段，在获得对应时间段内的客户的微表情走势，即，获得每个销售话术下产品的微表情走势，从而帮助业务人员快速掌握在销售话术下客户的产品瞬间认可度变化趋势。
步骤S523，根据每个销售话术下产品的微表情走势获得关键话术，并将所述关键话术添加至智能话术库。
在获取每个销售话术下产品的微表情走势后，将微表情走势上升的时间段内对应的销售话术作为关键话术，即，将客户的产品瞬间认可度上升的时间段内对应的销售话术作为关键话术，再将关键话术添加至智能话术库。将能够提升客户认可程度的关键话术添加至智能话术库，使得业务人员可以根据智能话术库中的关键话术来及时调整销售话术和服务质量，以此达成销售意向。即，通过智能话术库来反哺业务人员自身话术，从而提升业务人员的销售水平，节约业务人员话术培训投入成本，也去除客户回访大量人力物力成本。
在其中一个实施例中，请参照图6，图6为本申请产品认可度分析方法第三实施例的流程示意图，基于上述实施例，步骤S530之后，还包括：
步骤S540，获取产品介绍时间段内所有客户对同一产品的认可程度等级；
具体地，产品认可度等级分为很高、较高、较低以及很低，通过微表情走势图获取单个客户对产品的认可程度等级后，终端可以获取产品介绍时间段内的所有客户对同一产品的认可度等级，例如，当介绍的产品有A、B、C三种时，获取业务人员进行了推销的所有的客户对A的认可度等级，对B的认可度等级以及对C的认可度等级。
步骤S550，统计不同的产品认可度等级的客户数；
具体地，在获取到所有客户对同一产品的认可度等级后，统计不同的产品认可度等级下的客户数，即，统计产品认可度等级分别为很高、较高、较低、很低时，每个产品认可度等级下的客户数。
步骤S560，根据所述客户数获得市场对同一产品的认可程度。
具体地，在获得不同的产品认可度等级下的客户数以后，根据每个产品认可度等级下的客户数可以判断整个市场对该产品的认可程度，如果产品认可度等级为很高的客户数最多，说明整个市场对产品的认可度很高；如果产品认可度等级为较高的客户数最多，说明整个市场对产品的认可度较高；如果产品认可度等级为较低的客户数最多，说明整个市场对产品的认可度较低；如果产品认可度等级为很低的客户数最多，说明整个市场对产品的认可度很低。
在其中一个实施例中，请参照图7，图7为本申请产品认可度分析方法第四实施例的流程示意图，基于第一实施例，步骤S530之后，还包括：
步骤S570，获取产品介绍时间段内单个客户对不同产品的认可程度等级；
具体地，可以对业务人员给客户介绍产品时的视频图像进行识别获得客户的产品瞬间认可度，因此，可以对同一客户被介绍不同产品的视频图像进 行识别，在同一客户被介绍不同产品的视频图像全部识别完毕后，可以获取单个客户对不同产品的认可程度等级。
步骤S580，根据单个客户对不同产品的认可程度等级绘制出单个客户对不同产品的偏好分布图；
在获取到单个客户对不同产品的认可程度等级后，根据单个客户对不同产品的认可程度等级绘制出单个客户对不同产品的偏好分布图，分布图的横坐标为产品的类型，纵坐标为产品认可度等级。
步骤S590，根据所述偏好分布图获取客户的产品偏好。
具体地，在绘制完偏好分布图后，在偏好分布图中，认可程度等级越高，说明客户对该认可程度等级对应的产品越认可，因此，可以通过产品偏好图直接得到客户的产品偏好属性。
此外，在另一种实施例中，还可以获取所有客户对不同产品的产品认可度等级，将所有客户按照年龄进行区分，获取不同年龄的客户对不同产品的产品认可度等级。例如，针对A产品，得到不同年龄的客户对A产品的产品认可度等级，统计不同产品认可度等级下客户的年龄分布情况，对A产品的产品认可度等级为很高的客户中哪个年龄段的客户的比例最多，说明A产品比较受该年龄段的喜欢。
在其中一个实施例中，在建立产品认可度微表情模型时，可以先采集微表情图像样本，将微表情图像样本通过Gamma微表情接口，输入到产品认可度微表情模型，在原始微表情识别模型对微表情图像样本进行特征提取、聚类并设置相关系数，生成投资意愿度公式。同时，在有数据时，通过卷积神经网络训练客户对金融产品认可度的产品认可度微表情模型，不断纠正公式，不断提高准确度。模型公式的纠正过程是基于实际进行产品购买的的用户的产品认可微表情值数据(说明产品认可度很高)和未购买产品的客户(说明对产品认可度很低)的产品认可微表情值数据和表情图片，获取到返回的实际微表情类型对应的瞬时置信率，然后对公式中的相关系数进行调整优化反哺，提高模型的准确度。
此外，请参照图8，本申请还提供一种产品认可度分析装置10，所述产品认可度分析装置10包括：
第一获取模,20，所述第一获取模块用于获取产品介绍时间段内的视频图像；
识别模块30，所述识别模块用于对所述视频图像进行识别获取所述视频图像中人脸微表情对应的微表情类型以及所述微表情类型对应的瞬时置信度；
第二获取模块40，所述第二获取模块用于根据所述微表情类型获取对应的相关系数；
计算模块50，所述计算模块用于将所述瞬时置信度和所述相关系数填入投资意愿度公式，得到产品瞬间认可度；
第三获取模块60，所述第三获取模块用于根据产品介绍时间段内的产品 瞬间认可度获得客户对产品的认可程度。
在其中一个实施例中，所述第三获取模块60还用于：
获取产品介绍时间段内单个客户对产品的产品瞬间认可度；
根据所述产品瞬间认可度绘制出产品介绍时间段内单个客户对产品的微表情走势图；
根据所述微表情走势图获取单个客户对产品的认可程度等级。
在其中一个实施例中，所述产品认可度分析装置10还包括：
第四获取模块，所述第四获取模块用于获取产品介绍时间段内的销售话术；
第五获取模块，所述第五获取模块用于根据产品的微表情走势图获得每个销售话术下产品的微表情走势；
添加模块，所述添加模块用于根据每个销售话术下产品的微表情走势获得关键话术，并将所述关键话术添加至智能话术库。
在其中一个实施例中，所述产品认可度分析装置10还包括：
第六获取模块，所述第六获取模块用于获取产品介绍时间段内所有客户对同一产品的认可程度等级；
统计模块，所述统计模块用于统计不同的产品认可度等级的客户数；
第七获取模块，所述第七获取模块用于根据所述客户数获得市场对同一产品的认可程度。
在其中一个实施例中，所述产品认可度分析装置10还包括：
第八获取模块，所述第八获取模块用于获取产品介绍时间段内单个客户对不同产品的认可程度等级；
绘制模块，所述绘制模块用于根据单个客户对不同产品的认可程度等级绘制出单个客户对不同产品的偏好分布图；
第九获取模块，所述第九获取模块用于根据所述偏好分布图获取客户的产品偏好。
在其中一个实施例中，所述识别模块30还用于：
对所述视频图像进行识别获取所述视频图像中人脸微表情对应的微表情特征；
根据所述微表情特征获得对应的微表情类型以及所述微表情类型对应的瞬时置信度。
其中，上述产品认可度分析装置10中各个模块与上述产品认可度分析方法实施例中各步骤相对应，其功能和实现过程在此处不再一一赘述。
此外，本申请还提供一种计算机可读存储介质，计算机可读存储介质可以为非易失性存储介质，也可以是易失性存储介质。本申请计算机可读存储介质上存储有产品认可度分析程序，其中，产品认可度分析程序被处理器执行时，实现如上述的产品认可度分析方法的步骤。
本领域内的技术人员应明白，本申请的实施例可提供为方法、系统、或计算机程序产品。因此，本申请可采用完全硬件实施例、完全软件实施例、 或结合软件和硬件方面的实施例的形式。而且，本申请可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
以上所述实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本申请一些示例性实施例，其中描述较为具体和详细，但并不能因此而理解为对本申请专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本申请构思的前提下，还可以做出若干变形和改进，这些都属于本申请的保护范围。因此，本申请专利的保护范围应以所附权利要求为准。

Claims (20)

  1. 一种产品认可度分析方法，所述产品认可度分析方法包括：
    获取产品介绍时间段内的视频图像；
    对所述视频图像进行识别获取所述视频图像中人脸微表情对应的微表情类型以及所述微表情类型对应的瞬时置信度；
    根据所述微表情类型获取对应的相关系数；
    将所述瞬时置信度和所述相关系数填入投资意愿度公式，得到产品瞬间认可度；
    根据产品介绍时间段内的产品瞬间认可度获得客户对产品的认可程度。
  2. 如权利要求1所述的产品认可度分析方法，其中，所述投资意愿度公式为
    Figure PCTCN2020086166-appb-100001
    其中，p i是指微表情类型i对应的相关系数，q i是指微表情类型i对应的瞬间瞬时置信度，R是指产品瞬间认可度。
  3. 如权利要求1所述的产品认可度分析方法，其中，所述根据产品介绍时间段内的产品瞬间认可度获得客户对产品的认可程度的步骤包括：
    获取产品介绍时间段内单个客户对产品的产品瞬间认可度；
    根据所述产品瞬间认可度绘制出产品介绍时间段内单个客户对产品的微表情走势图；
    根据所述微表情走势图获取单个客户对产品的认可程度等级。
  4. 如权利要求3所述的产品认可度分析方法，其中，所述根据所述产品瞬间认可度绘制出产品介绍时间段内单个客户对产品的微表情走势图的步骤之后，还包括：
    获取产品介绍时间段内的销售话术；
    根据产品的微表情走势图获得每个销售话术下产品的微表情走势；
    根据每个销售话术下产品的微表情走势获得关键话术，并将所述关键话术添加至智能话术库。
  5. 如权利要求3所述的产品认可度分析方法，其中，所述根据所述微表情走势图获取单个客户对产品的认可程度等级的步骤之后，还包括：
    获取产品介绍时间段内所有客户对同一产品的认可程度等级；
    统计不同的产品认可度等级的客户数；
    根据所述客户数获得市场对同一产品的认可程度。
  6. 如权利要求3所述的产品认可度分析方法，其中，所述根据所述微表情走势图获取单个客户对产品的认可程度等级的步骤之后，还包括：
    获取产品介绍时间段内单个客户对不同产品的认可程度等级；
    根据单个客户对不同产品的认可程度等级绘制出单个客户对不同产品的偏好分布图；
    根据所述偏好分布图获取客户的产品偏好。
  7. 如权利要求1-6中任一项所述的产品认可度分析方法，其中，所述对 所述视频图像进行识别获取所述视频图像中人脸微表情对应的微表情类型以及所述微表情类型对应的瞬时置信度的步骤包括：
    对所述视频图像进行识别获取所述视频图像中人脸微表情对应的微表情特征；
    根据所述微表情特征获得对应的微表情类型以及所述微表情类型对应的瞬时置信度。
  8. 一种产品认可度分析装置，所述产品认可度分析装置包括：
    第一获取模块，所述第一获取模块用于获取产品介绍时间段内的视频图像；
    识别模块，所述识别模块用于对所述视频图像进行识别获取所述视频图像中人脸微表情对应的微表情类型以及所述微表情类型对应的瞬时置信度；
    第二获取模块，所述第二获取模块用于根据所述微表情类型获取对应的相关系数；
    计算模块，所述计算模块用于将所述瞬时置信度和所述相关系数填入投资意愿度公式，得到产品瞬间认可度；
    第三获取模块，所述第三获取模块用于根据产品介绍时间段内的产品瞬间认可度获得客户对产品的认可程度。
  9. 一种终端，包括处理器、存储器、以及存储在所述存储器上的可被所述处理器执行的产品认可度分析程序，其中，所述产品认可度分析程序被所述处理器执行时，实现以下方法步骤：
    获取产品介绍时间段内的视频图像；
    对所述视频图像进行识别获取所述视频图像中人脸微表情对应的微表情类型以及所述微表情类型对应的瞬时置信度；
    根据所述微表情类型获取对应的相关系数；
    将所述瞬时置信度和所述相关系数填入投资意愿度公式，得到产品瞬间认可度；
    根据产品介绍时间段内的产品瞬间认可度获得客户对产品的认可程度。
  10. 如权利要求9所述的产品认可度分析方法，其中，所述投资意愿度公式为
    Figure PCTCN2020086166-appb-100002
    其中，p i是指微表情类型i对应的相关系数，q i是指微表情类型i对应的瞬间瞬时置信度，R是指产品瞬间认可度。
  11. 如权利要求9所述终端，所述根据产品介绍时间段内的产品瞬间认可度获得客户对产品的认可程度的步骤包括：
    获取产品介绍时间段内单个客户对产品的产品瞬间认可度；
    根据所述产品瞬间认可度绘制出产品介绍时间段内单个客户对产品的微表情走势图；
    根据所述微表情走势图获取单个客户对产品的认可程度等级。
  12. 如权利要求11所述终端，其中，所述根据所述产品瞬间认可度绘制出产品介绍时间段内单个客户对产品的微表情走势图的步骤之后，还包括：
    获取产品介绍时间段内的销售话术；
    根据产品的微表情走势图获得每个销售话术下产品的微表情走势；
    根据每个销售话术下产品的微表情走势获得关键话术，并将所述关键话术添加至智能话术库。
  13. 如权利要求11所述终端，其中，所述根据所述微表情走势图获取单个客户对产品的认可程度等级的步骤之后，还包括：
    获取产品介绍时间段内所有客户对同一产品的认可程度等级；
    统计不同的产品认可度等级的客户数；
    根据所述客户数获得市场对同一产品的认可程度。
  14. 如权利要求12所述终端，其中，所述根据所述微表情走势图获取单个客户对产品的认可程度等级的步骤之后，还包括：
    获取产品介绍时间段内单个客户对不同产品的认可程度等级；
    根据单个客户对不同产品的认可程度等级绘制出单个客户对不同产品的偏好分布图；
    根据所述偏好分布图获取客户的产品偏好。
  15. 如权利要求9-14中任一项所述终端，其中，所述对所述视频图像进行识别获取所述视频图像中人脸微表情对应的微表情类型以及所述微表情类型对应的瞬时置信度的步骤包括：
    对所述视频图像进行识别获取所述视频图像中人脸微表情对应的微表情特征；
    根据所述微表情特征获得对应的微表情类型以及所述微表情类型对应的瞬时置信度。
  16. 一种计算机可读存储介质，所述计算机可读存储介质上存储有产品认可度分析程序，其中，所述产品认可度分析程序被处理器执行时，实现以下方法的步骤：
    获取产品介绍时间段内的视频图像；
    对所述视频图像进行识别获取所述视频图像中人脸微表情对应的微表情类型以及所述微表情类型对应的瞬时置信度；
    根据所述微表情类型获取对应的相关系数；
    将所述瞬时置信度和所述相关系数填入投资意愿度公式，得到产品瞬间认可度；
    根据产品介绍时间段内的产品瞬间认可度获得客户对产品的认可程度。
  17. 如权利要求16所述计算机可读存储介质，其中，所述投资意愿度公式为
    Figure PCTCN2020086166-appb-100003
    其中，p i是指微表情类型i对应的相关系数，q i是指微表情类型i对应的瞬间瞬时置信度，R是指产品瞬间认可度。
  18. 如权利要求16所述计算机可读存储介质，其中，所述根据产品介绍时间段内的产品瞬间认可度获得客户对产品的认可程度的步骤包括：
    获取产品介绍时间段内单个客户对产品的产品瞬间认可度；
    根据所述产品瞬间认可度绘制出产品介绍时间段内单个客户对产品的微表情走势图；
    根据所述微表情走势图获取单个客户对产品的认可程度等级。
  19. 如权利要求18所述计算机可读存储介质，其中，所述根据所述产品瞬间认可度绘制出产品介绍时间段内单个客户对产品的微表情走势图的步骤之后，还包括：
    获取产品介绍时间段内的销售话术；
    根据产品的微表情走势图获得每个销售话术下产品的微表情走势；
    根据每个销售话术下产品的微表情走势获得关键话术，并将所述关键话术添加至智能话术库。
  20. 如权利要求18所述计算机可读存储介质，其中，所述根据所述微表情走势图获取单个客户对产品的认可程度等级的步骤之后，还包括：
    获取产品介绍时间段内所有客户对同一产品的认可程度等级；
    统计不同的产品认可度等级的客户数；
    根据所述客户数获得市场对同一产品的认可程度。
PCT/CN2020/086166 2019-06-19 2020-04-22 产品认可度分析方法、装置、终端及计算机可读存储介质 Ceased WO2020253363A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910532837.3A CN110415015A (zh) 2019-06-19 2019-06-19 产品认可度分析方法、装置、终端及计算机可读存储介质
CN201910532837.3 2019-06-19

Publications (1)

Publication Number Publication Date
WO2020253363A1 true WO2020253363A1 (zh) 2020-12-24

Family

ID=68359333

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/086166 Ceased WO2020253363A1 (zh) 2019-06-19 2020-04-22 产品认可度分析方法、装置、终端及计算机可读存储介质

Country Status (2)

Country Link
CN (1) CN110415015A (zh)
WO (1) WO2020253363A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110415015A (zh) * 2019-06-19 2019-11-05 深圳壹账通智能科技有限公司 产品认可度分析方法、装置、终端及计算机可读存储介质
CN114037201A (zh) * 2021-09-29 2022-02-11 国家电网有限公司 一种营业厅服务效能评估方法及系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190050633A1 (en) * 2016-06-15 2019-02-14 Stephan Hau Computer-based micro-expression analysis
CN109784312A (zh) * 2019-02-18 2019-05-21 深圳锐取信息技术股份有限公司 教学管理方法及装置
CN109858958A (zh) * 2019-01-17 2019-06-07 深圳壹账通智能科技有限公司 基于微表情的目标客户定位方法、装置、设备及存储介质
CN110415015A (zh) * 2019-06-19 2019-11-05 深圳壹账通智能科技有限公司 产品认可度分析方法、装置、终端及计算机可读存储介质

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107480622A (zh) * 2017-08-07 2017-12-15 深圳市科迈爱康科技有限公司 微表情识别方法、装置及存储介质
CN109858405A (zh) * 2019-01-17 2019-06-07 深圳壹账通智能科技有限公司 基于微表情的满意度评价方法、装置、设备及存储介质

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190050633A1 (en) * 2016-06-15 2019-02-14 Stephan Hau Computer-based micro-expression analysis
CN109858958A (zh) * 2019-01-17 2019-06-07 深圳壹账通智能科技有限公司 基于微表情的目标客户定位方法、装置、设备及存储介质
CN109784312A (zh) * 2019-02-18 2019-05-21 深圳锐取信息技术股份有限公司 教学管理方法及装置
CN110415015A (zh) * 2019-06-19 2019-11-05 深圳壹账通智能科技有限公司 产品认可度分析方法、装置、终端及计算机可读存储介质

Also Published As

Publication number Publication date
CN110415015A (zh) 2019-11-05

Similar Documents

Publication Publication Date Title
CN112733042B (zh) 推荐信息的生成方法、相关装置及计算机程序产品
CN111428010B (zh) 人机智能问答的方法和装置
CN107481720B (zh) 一种显式声纹识别方法及装置
CN112417158A (zh) 文本数据分类模型的训练方法、分类方法、装置和设备
CN108280200B (zh) 用于推送信息的方法和装置
WO2020088058A1 (zh) 生成信息的方法和装置
CN107886949A (zh) 一种内容推荐方法及装置
CN107193974B (zh) 基于人工智能的地域性信息确定方法和装置
CN111353028A (zh) 用于确定客服话术簇的方法及装置
CN112036923A (zh) 一种服务评价方法、系统、装置及存储介质
WO2023272616A1 (zh) 一种文本理解方法、系统、终端设备和存储介质
CN114138960A (zh) 用户意图识别方法、装置、设备及介质
CN109582825B (zh) 用于生成信息的方法和装置
US10692498B2 (en) Question urgency in QA system with visual representation in three dimensional space
CN110399465A (zh) 用于处理信息的方法和装置
CN116204624A (zh) 应答方法、装置、电子设备及存储介质
CN115640386A (zh) 用于基于推荐话术进行对话的方法和设备
CN114691850A (zh) 生成问答对的方法、神经网络模型的训练方法和装置
CN108287848B (zh) 用于语义解析的方法和系统
CN110009480A (zh) 司法催收路径的推荐方法、装置、介质、电子设备
CN112115248A (zh) 一种从对话语料中抽取对话策略结构的方法及系统
WO2020253363A1 (zh) 产品认可度分析方法、装置、终端及计算机可读存储介质
CN112131378B (zh) 用于识别民生问题类别的方法、装置及电子设备
CN120561236A (zh) 智能问答方法、设备、存储介质及计算机程序产品
CN117454015B (zh) 一种信息推荐方法及装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20826353

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20826353

Country of ref document: EP

Kind code of ref document: A1