WO2019061990A1 - 用户意图预测方法、电子设备及计算机可读存储介质 - Google Patents

用户意图预测方法、电子设备及计算机可读存储介质 Download PDF

Info

Publication number
WO2019061990A1
WO2019061990A1 PCT/CN2018/076177 CN2018076177W WO2019061990A1 WO 2019061990 A1 WO2019061990 A1 WO 2019061990A1 CN 2018076177 W CN2018076177 W CN 2018076177W WO 2019061990 A1 WO2019061990 A1 WO 2019061990A1
Authority
WO
WIPO (PCT)
Prior art keywords
user
classification
specific user
specific
incoming
Prior art date
Application number
PCT/CN2018/076177
Other languages
English (en)
French (fr)
Inventor
安欣
王建明
肖京
Original Assignee
平安科技（深圳）有限公司
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 平安科技（深圳）有限公司 filed Critical 平安科技（深圳）有限公司
Publication of WO2019061990A1 publication Critical patent/WO2019061990A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/25—Fusion techniques
    • G06F18/254—Fusion techniques of classification results, e.g. of results related to same input data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/01—Customer relationship services
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L13/00—Speech synthesis; Text to speech systems
    • G10L13/08—Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue

Definitions

  • the present application relates to the field of computer information technology, and in particular, to a user intention prediction method, an electronic device, and a computer readable storage medium.
  • the present application proposes a user intention prediction method, an electronic device, and a computer readable storage medium.
  • voice recognition technology By combining voice recognition technology with big data processing technology, the efficiency and accuracy of user intention prediction are improved.
  • the present application provides an electronic device including a memory, a processor, and a user intent prediction system stored on the memory and operable on the processor, the user intent
  • the prediction system is implemented by the processor to implement the following steps:
  • Predetermined classification algorithm is used to perform predetermined number of classification modeling on user characteristics and incoming flow characteristics of a specific user to obtain each classification model, and obtain each classification probability returned by each classification model;
  • the classification with the highest probability is selected to predict the business intention of the specific user.
  • the user characteristics include a natural attribute, a group attribute, a social attribute, and a contact attribute;
  • the natural attribute includes age, region, gender, and education
  • the group attribute includes a product attribute of a product purchased by a user in a group
  • the social attribute includes a profession, a child, and a job company
  • the contact attribute includes the user in a group.
  • the incoming flow feature includes the user triggering a particular service or product by a particular incoming mode, wherein the incoming flow features are sorted according to chronological order.
  • each classification node is modeled by a different feature importance
  • Each of the classification nodes is modeled by different feature importance including:
  • the feature importance of each classification is sorted from high to low, and specific features in each classification are selected in parallel as important features, and modeling is performed according to the important features selected in each classification.
  • the user intent prediction system is further used to implement the following steps when executed by the processor:
  • the text content of the predicted result is converted into voice content by a preset voice recognition algorithm, and the voice content is broadcast to the user end corresponding to the specific user.
  • the present application further provides a user intention prediction method, which is applied to an electronic device, and the method includes:
  • Predetermined classification algorithm is used to perform predetermined number of classification modeling on user characteristics and incoming flow characteristics of a specific user to obtain each classification model, and obtain each classification probability returned by each classification model;
  • the classification with the highest probability is selected to predict the business intention of the specific user.
  • the user features include a natural attribute, a group attribute, a social attribute, and a contact attribute;
  • the natural attribute includes age, region, gender, and education
  • the group attribute includes a product attribute of a product purchased by a user in a group
  • the social attribute includes a profession, a child, and a job company
  • the contact attribute includes the user in a group.
  • the incoming pipeline feature includes a user triggering a particular service or product through a particular incoming mode, wherein the incoming pipeline features are sorted according to chronological order.
  • each classification node is modeled by a different feature importance
  • Each of the classification nodes is modeled by different feature importance including:
  • the feature importance of each classification is sorted from high to low, and specific features in each classification are selected in parallel as important features, and modeling is performed according to the important features selected in each classification.
  • the user intention prediction method further comprises the steps of:
  • the text content of the predicted result is converted into voice content by a preset voice recognition algorithm, and the voice content is broadcast to the user end corresponding to the specific user.
  • the present application further provides a computer readable storage medium storing a user intent prediction system, the user intent prediction system being executable by at least one processor, such that The at least one processor performs the steps of the user intent prediction method as described above.
  • the electronic device, the user intention prediction method and the computer readable storage medium proposed by the present application improve the efficiency and accuracy of the user's intention prediction by combining the voice recognition technology and the big data processing technology.
  • Applying to the enterprise customer service (such as customer phone service, etc.) can greatly shorten the time for users to handle business through customer service calls and improve user satisfaction.
  • 1 is a schematic diagram of an optional hardware architecture of an electronic device of the present application
  • FIG. 2 is a schematic diagram of a program module of an embodiment of a user intention prediction system in an electronic device of the present application
  • FIG. 3 is a schematic diagram of an implementation process of an embodiment of a user intention prediction method according to the present application.
  • first, second and the like in the present application are for the purpose of description only, and are not to be construed as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. .
  • features defining “first” and “second” may include at least one of the features, either explicitly or implicitly.
  • the technical solutions between the various embodiments may be combined with each other, but must be based on the realization of those skilled in the art, and when the combination of the technical solutions is contradictory or impossible to implement, it should be considered that the combination of the technical solutions does not exist. Nor is it within the scope of protection required by this application.
  • FIG. 1 it is a schematic diagram of an optional hardware architecture of the electronic device 2 of the present application.
  • the electronic device 2 may include, but is not limited to, a memory 21, a processor 22, and a network interface 23 that can communicate with each other through a system bus. It is pointed out that FIG. 1 only shows the electronic device 2 with the components 21-23, but it should be understood that not all illustrated components are required to be implemented, and more or fewer components may be implemented instead.
  • the electronic device 2 may be a computing device such as a rack server, a blade server, a tower server, or a rack server.
  • the electronic device 2 may be an independent server or a server cluster composed of multiple servers. .
  • the memory 21 includes at least one type of readable storage medium including a flash memory, a hard disk, a multimedia card, a card type memory (eg, SD or DX memory, etc.), a random access memory (RAM), a static Random access memory (SRAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), programmable read only memory (PROM), magnetic memory, magnetic disk, optical disk, and the like.
  • the memory 21 may be an internal storage unit of the electronic device 2, such as a hard disk or memory of the electronic device 2.
  • the memory 21 may also be an external storage device of the electronic device 2, such as a plug-in hard disk equipped on the electronic device 2, a smart memory card (SMC), and a secure digital device. (Secure Digital, SD) card, flash card, etc.
  • the memory 21 may also include both an internal storage unit of the electronic device 2 and an external storage device thereof.
  • the memory 21 is generally used to store an operating system installed in the electronic device 2 and various types of application software, such as program codes of the user intention prediction system 20, and the like. Further, the memory 21 can also be used to temporarily store various types of data that have been output or are to be output.
  • the processor 22 may be a Central Processing Unit (CPU), controller, microcontroller, microprocessor, or other data processing chip in some embodiments.
  • the processor 22 is typically used to control the overall operation of the electronic device 2, such as performing control and processing related to data interaction or communication with the electronic device 2.
  • the processor 22 is configured to run program code or process data stored in the memory 21, such as running the user intention prediction system 20 and the like.
  • the network interface 23 may comprise a wireless network interface or a wired network interface, which is typically used to establish a communication connection between the electronic device 2 and other electronic devices.
  • the network interface 23 is configured to connect the electronic device 2 to an external data platform through a network, and establish a data transmission channel and a communication connection between the electronic device 2 and an external data platform.
  • the network may be an intranet, an Internet, a Global System of Mobile communication (GSM), a Wideband Code Division Multiple Access (WCDMA), a 4G network, or a 5G network.
  • Wireless or wired networks such as network, Bluetooth, Wi-Fi, etc.
  • FIG. 2 it is a program module diagram of an embodiment of the user intention prediction system 20 in the electronic device 2 of the present application.
  • the user intent prediction system 20 may be divided into one or more program modules, the one or more program modules being stored in the memory 21 and being processed by one or more processors ( This embodiment is executed by the processor 22) to complete the application.
  • the user intent prediction system 20 can be segmented into an establishment module 201, an acquisition module 202, and a prediction module 203.
  • a program module as referred to in this application refers to a series of computer program instructions that are capable of performing a particular function, and are more suitable than the program to describe the execution of the user intent prediction system 20 in the electronic device 2. The function of each program module 201-203 will be described in detail below.
  • the establishing module 201 is configured to establish a user feature and an incoming pipeline feature.
  • the user features include, but are not limited to, a natural attribute, a group attribute, a social attribute, a contact attribute (a business subsidiary), and the like.
  • the natural attribute may include the following information: age, region, gender, education, etc.
  • the group attribute may include the following information: product attributes of the product purchased by the user in a group, such as a property insurance user, etc.
  • the information may include the following: occupation, child, employment company, etc.
  • the contact attribute may include the following information: the product attribute of the product purchased by the user in a group business subsidiary, such as the property insurance specific purchase purchased by the user in a group business subsidiary. The number of products and specific products purchased.
  • the incoming pipeline feature includes, but is not limited to, a user triggering a specific service or product through a specific incoming mode (such as a voice call incoming mode), such as consulting a specific service by phone, purchasing a specific product. Car insurance report, etc.
  • a specific incoming mode such as a voice call incoming mode
  • the inflow water flow characteristics may be classified according to a chronological order (such as year/month/week/day), for example, a specific product purchased by the user in the most recent year, the number of car insurance reports in the day, and the like.
  • the obtaining module 202 is configured to perform, by using a predetermined classification algorithm (such as a multi-classification algorithm combining Spark and Softmax), a predetermined number (such as 600 classes) for a user characteristic of a specific user and an incoming pipeline feature (such as 11000 dimensions). Modeling each classification model and obtaining the classification probabilities returned by each classification model.
  • a predetermined classification algorithm such as a multi-classification algorithm combining Spark and Softmax
  • a predetermined number such as 600 classes
  • an incoming pipeline feature such as 11000 dimensions
  • each classification node is modeled by a different feature importance.
  • important features include whether the specific user purchases car insurance, driving age, and voice telephone incoming report data in the day, and models and judges the specific user according to the above important features. The probability of reporting a car insurance is carried out to effectively improve the accuracy of the forecast.
  • the modeling of each classification node by different feature importance includes: for all predetermined quantities (600 categories) of classification, according to the characteristic importance of each classification from high to low Sorting, selecting specific features in each category (such as the top three features) in parallel as important features, modeling according to the important features selected in each category. Due to the feature importance analysis while modeling, the efficiency and accuracy of business forecasting are effectively improved.
  • the prediction module 203 is configured to sort according to each classification probability returned by each classification model, and select the classification with the highest probability to predict the business intention of the specific user. For example, if the classification with the highest probability of return is the property insurance business, it is predicted that the specific user has the intention to purchase property insurance.
  • the prediction module 203 is further configured to:
  • the preferred recommendation information of the specific user is obtained by the repeated incoming rate of the specific user within a predetermined time (for example, within 24 hours), thereby achieving accurate recommendation. For example, if the user who has reported the car insurance report within 24 hours enters the line twice, it is determined that the user has a higher probability to continue reporting, and therefore, the predicted result of the car insurance report is recommended as the preferred recommendation information of the user.
  • the prediction module 203 is further configured to:
  • the voice content can be set to: "Do you need to handle xx business (such as property insurance business)?".
  • the user intention prediction system 20 proposed by the present application improves the efficiency and accuracy of the user's intention prediction by combining the voice recognition technology and the big data processing technology, and applies the application to the enterprise customer service (such as customer telephone business, etc., can greatly shorten the time for users to handle business through customer service calls, and improve user satisfaction.
  • the enterprise customer service Such as customer telephone business, etc.
  • the present application also proposes a user intention prediction method.
  • FIG. 3 it is a schematic flowchart of an implementation process of an embodiment of the user intention prediction method of the present application.
  • the order of execution of the steps in the flowchart shown in FIG. 3 may be changed according to different requirements, and some steps may be omitted.
  • Step S31 establishing user characteristics and incoming flow characteristics.
  • the user features include, but are not limited to, a natural attribute, a group attribute, a social attribute, a contact attribute (a business subsidiary), and the like.
  • the natural attribute may include the following information: age, region, gender, education, etc.
  • the group attribute may include the following information: product attributes of the product purchased by the user in a group, such as a property insurance user, etc.
  • the information may include the following: occupation, child, employment company, etc.
  • the contact attribute may include the following information: the product attribute of the product purchased by the user in a group business subsidiary, such as the property insurance specific purchase purchased by the user in a group business subsidiary. The number of products and specific products purchased.
  • the incoming pipeline feature includes, but is not limited to, a user triggering a specific service or product through a specific incoming mode (such as a voice call incoming mode), such as consulting a specific service by phone, purchasing a specific product. Car insurance report, etc.
  • a specific incoming mode such as a voice call incoming mode
  • the inflow water flow characteristics may be classified according to a chronological order (such as year/month/week/day), for example, a specific product purchased by the user in the most recent year, the number of car insurance reports in the day, and the like.
  • a predetermined classification algorithm such as Spark and Softmax combined multi-classification algorithm
  • a predetermined number such as 600 classes
  • incoming flow characteristics such as 11000 dimensions
  • each classification node is modeled by a different feature importance.
  • important features include whether the specific user purchases car insurance, driving age, and voice telephone incoming report data in the day, and models and judges the specific user according to the above important features. The probability of reporting a car insurance is carried out to effectively improve the accuracy of the forecast.
  • the modeling of each classification node by different feature importance includes: for all predetermined quantities (600 categories) of classification, according to the characteristic importance of each classification from high to low Sorting, selecting specific features in each category (such as the top three features) in parallel as important features, modeling according to the important features selected in each category. Due to the feature importance analysis while modeling, the efficiency and accuracy of business forecasting are effectively improved.
  • Step S33 sorting according to each classification probability returned by each classification model, and selecting the classification with the highest probability to predict the business intention of the specific user. For example, if the classification with the highest probability of return is the property insurance business, it is predicted that the specific user has the intention to purchase property insurance.
  • the user intention prediction method further includes the following steps:
  • the preferred recommendation information of the specific user is obtained by the repeated incoming rate of the specific user within a predetermined time (for example, within 24 hours), thereby achieving accurate recommendation. For example, if the user who has reported the car insurance report within 24 hours enters the line twice, it is determined that the user has a higher probability to continue reporting, and therefore, the predicted result of the car insurance report is recommended as the preferred recommendation information of the user.
  • the user intention prediction method further includes the following steps:
  • the voice content can be set to: "Do you need to handle xx business (such as property insurance business)?".
  • the user intention prediction method proposed by the present application improves the efficiency and accuracy of the user's intention prediction by combining the speech recognition technology and the big data processing technology, and applies the application to the enterprise customer service.
  • Business such as customer telephone service, etc.
  • the present application further provides a computer readable storage medium (such as a ROM/RAM, a magnetic disk, an optical disk) storing a user intention prediction system 20, the user intention
  • the prediction system 20 can be executed by at least one processor 22 to cause the at least one processor 22 to perform the steps of the user intent prediction method as described above.
  • the foregoing embodiment method can be implemented by means of software plus a necessary general hardware platform, and can also be implemented by hardware, but in many cases, the former is A better implementation.
  • the technical solution of the present application which is essential or contributes to the prior art, may be embodied in the form of a software product stored in a storage medium (such as ROM/RAM, disk,
  • the optical disc includes a number of instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the methods described in various embodiments of the present application.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Human Computer Interaction (AREA)
  • Development Economics (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Human Resources & Organizations (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Finance (AREA)
  • Game Theory and Decision Science (AREA)
  • Accounting & Taxation (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

本申请公开了一种用户意图预测方法，该方法包括步骤：建立用户特征和进线流水特征；通过预定分类算法，针对特定用户的用户特征和进线流水特征进行预定数量的分类建模得到各分类模型，并获取各分类模型返回的各分类概率；依据各分类模型返回的各分类概率进行排序，选取概率最高的分类预测该特定用户的业务意图。本申请可以提高用户意图预测的效率和准确度。

Description

用户意图预测方法、电子设备及计算机可读存储介质
本申请要求于2017年09月30日提交中国专利局、申请号为201710939717.6、发明名称为“用户意图预测方法、电子设备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申请中。
技术领域
本申请涉及计算机信息技术领域，尤其涉及一种用户意图预测方法、电子设备及计算机可读存储介质。
背景技术
目前，用户通过企业客服电话中的语音导航办理业务步骤复杂，耗时过长。原因在于业务节点太过密集，特征维数过多。因此，如何缩短用户通过客服电话办理业务的时间是急需解决的技术问题。
发明内容
有鉴于此，本申请提出一种用户意图预测方法、电子设备及计算机可读存储介质，通过结合语音识别技术与大数据处理技术，提高了用户意图预测的效率和准确度。
首先，为实现上述目的，本申请提出一种电子设备，所述电子设备包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的用户意图预测系统，所述用户意图预测系统被所述处理器执行时实现如下步骤：
建立用户特征和进线流水特征；
通过预定分类算法，针对特定用户的用户特征和进线流水特征进行预定数量的分类建模得到各分类模型，并获取各分类模型返回的各分类概率；及
依据各分类模型返回的各分类概率进行排序，选取概率最高的分类预测 该特定用户的业务意图。
优选地，所述用户特征包括自然属性、集团属性、社会属性、接触属性；及
所述自然属性包括年龄、地区、性别、学历，所述集团属性包括用户在某集团所购买产品的产品属性，所述社会属性包括职业、子女、就职公司，所述接触属性包括用户在某集团业务子公司所购买产品的产品属性。
优选地，所述进线流水特征包括用户通过特定进线方式触发特定业务或产品，其中，所述进线流水特征根据时间顺序进行分类。
优选地，每个分类节点通过不同的特征重要性进行建模；及
所述每个分类节点通过不同的特征重要性进行建模包括：
针对所有预定数量的分类，按照每个分类的特征重要性由高至低进行排序，并行选取每个分类中的特定特征作为重要特征，根据每个分类中选取的重要特征进行建模。
优选地，所述用户意图预测系统被所述处理器执行时还用于实现如下步骤：
通过预定时间内该特定用户的反复进线率，获取该特定用户的优选推荐信息；及
通过预设的语音识别算法，将预测结果的文本内容转换为语音内容，并将该语音内容播报至该特定用户对应的用户端。
此外，为实现上述目的，本申请还提供一种用户意图预测方法，该方法应用于电子设备，所述方法包括：
建立用户特征和进线流水特征；
通过预定分类算法，针对特定用户的用户特征和进线流水特征进行预定数量的分类建模得到各分类模型，并获取各分类模型返回的各分类概率；及
依据各分类模型返回的各分类概率进行排序，选取概率最高的分类预测 该特定用户的业务意图。
优选地，所述用户特征包括自然属性、集团属性、社会属性、接触属性；
所述自然属性包括年龄、地区、性别、学历，所述集团属性包括用户在某集团所购买产品的产品属性，所述社会属性包括职业、子女、就职公司，所述接触属性包括用户在某集团业务子公司所购买产品的产品属性；及
所述进线流水特征包括用户通过特定进线方式触发特定业务或产品，其中，所述进线流水特征根据时间顺序进行分类。
优选地，每个分类节点通过不同的特征重要性进行建模；及
所述每个分类节点通过不同的特征重要性进行建模包括：
针对所有预定数量的分类，按照每个分类的特征重要性由高至低进行排序，并行选取每个分类中的特定特征作为重要特征，根据每个分类中选取的重要特征进行建模。
优选地，所述用户意图预测方法还包括步骤：
通过预定时间内该特定用户的反复进线率，获取该特定用户的优选推荐信息；及
通过预设的语音识别算法，将预测结果的文本内容转换为语音内容，并将该语音内容播报至该特定用户对应的用户端。
进一步地，为实现上述目的，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存储有用户意图预测系统，所述用户意图预测系统可被至少一个处理器执行，以使所述至少一个处理器执行如上述的用户意图预测方法的步骤。
相较于现有技术，本申请所提出的电子设备、用户意图预测方法及计算机可读存储介质，通过结合语音识别技术与大数据处理技术，提高了用户意图预测的效率和准确度，将本申请应用于企业客服业务(如客户电话业务等)， 可以大大缩短用户通过客服电话办理业务的时间，提高用户的满意度。
附图说明
图1是本申请电子设备一可选的硬件架构的示意图；
图2是本申请电子设备中用户意图预测系统一实施例的程序模块示意图；
图3为本申请用户意图预测方法一实施例的实施流程示意图。
附图标记：
电子设备 2
存储器 21
处理器 22
网络接口 23
用户意图预测系统 20
建立模块 201
获取模块 202
预测模块 203
流程步骤 S31-S33
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
为了使本申请的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处所描述的具体实施例仅用以解释本申请，并不用于限定本申请。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
需要说明的是，在本申请中涉及“第一”、“第二”等的描述仅用于描述目的，而不能理解为指示或暗示其相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或者隐含地包括至少一个该特征。另外，各个实施例之间的技术方案可以相互结合，但是必须是以本领域普通技术人员能够实现为基础，当技术方案的结合出现相互矛盾或无法实现时应当认为这种技术方案的结合不存在，也不在本申请要求的保护范围之内。
进一步需要说明的是，在本文中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者装置不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者装置所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括该要素的过程、方法、物品或者装置中还存在另外的相同要素。
首先，本申请提出一种电子设备2。
参阅图1所示，是本申请电子设备2一可选的硬件架构的示意图。本实施例中，所述电子设备2可包括，但不限于，可通过系统总线相互通信连接存储器21、处理器22、网络接口23。需要指出的是，图1仅示出了具有组件21-23的电子设备2，但是应理解的是，并不要求实施所有示出的组件，可以替代的实施更多或者更少的组件。
其中，所述电子设备2可以是机架式服务器、刀片式服务器、塔式服务器或机柜式服务器等计算设备，该电子设备2可以是独立的服务器，也可以是多个服务器所组成的服务器集群。
所述存储器21至少包括一种类型的可读存储介质，所述可读存储介质包括闪存、硬盘、多媒体卡、卡型存储器(例如，SD或DX存储器等)、随机访问存储器(RAM)、静态随机访问存储器(SRAM)、只读存储器(ROM)、 电可擦除可编程只读存储器(EEPROM)、可编程只读存储器(PROM)、磁性存储器、磁盘、光盘等。在一些实施例中，所述存储器21可以是所述电子设备2的内部存储单元，例如该电子设备2的硬盘或内存。在另一些实施例中，所述存储器21也可以是所述电子设备2的外部存储设备，例如该电子设备2上配备的插接式硬盘，智能存储卡(Smart Media Card,SMC)，安全数字(Secure Digital,SD)卡，闪存卡(Flash Card)等。当然，所述存储器21还可以既包括所述电子设备2的内部存储单元也包括其外部存储设备。本实施例中，所述存储器21通常用于存储安装于所述电子设备2的操作系统和各类应用软件，例如所述用户意图预测系统20的程序代码等。此外，所述存储器21还可以用于暂时地存储已经输出或者将要输出的各类数据。
所述处理器22在一些实施例中可以是中央处理器(Central Processing Unit，CPU)、控制器、微控制器、微处理器、或其他数据处理芯片。该处理器22通常用于控制所述电子设备2的总体操作，例如执行与所述电子设备2进行数据交互或者通信相关的控制和处理等。本实施例中，所述处理器22用于运行所述存储器21中存储的程序代码或者处理数据，例如运行所述的用户意图预测系统20等。
所述网络接口23可包括无线网络接口或有线网络接口，该网络接口23通常用于在所述电子设备2与其他电子设备之间建立通信连接。例如，所述网络接口23用于通过网络将所述电子设备2与外部数据平台相连，在所述电子设备2与外部数据平台之间建立数据传输通道和通信连接。所述网络可以是企业内部网(Intranet)、互联网(Internet)、全球移动通讯系统(Global System of Mobile communication，GSM)、宽带码分多址(Wideband Code Division Multiple Access，WCDMA)、4G网络、5G网络、蓝牙(Bluetooth)、Wi-Fi等无线或有线网络。
至此，己经详细介绍了本申请各个实施例的应用环境和相关设备的硬件结构和功能。下面，将基于上述应用环境和相关设备，提出本申请的各个实 施例。
参阅图2所示，是本申请电子设备2中用户意图预测系统20一实施例的程序模块图。本实施例中，所述的用户意图预测系统20可以被分割成一个或多个程序模块，所述一个或者多个程序模块被存储于所述存储器21中，并由一个或多个处理器(本实施例中为所述处理器22)所执行，以完成本申请。例如，在图2中，所述的用户意图预测系统20可以被分割成建立模块201、获取模块202、以及预测模块203。本申请所称的程序模块是指能够完成特定功能的一系列计算机程序指令段，比程序更适合于描述所述用户意图预测系统20在所述电子设备2中的执行过程。以下将就各程序模块201-203的功能进行详细描述。
所述建立模块201，用于建立用户特征和进线流水特征。
优选地，在本实施例中，所述用户特征包括，但不限于，自然属性、集团属性、社会属性、接触属性(业务子公司)等。其中，所述自然属性可以包括如下信息：年龄、地区、性别、学历等；所述集团属性可以包括如下信息：用户在某集团所购买产品的产品属性，如产险用户等；所述社会属性可以包括如下信息：职业、子女、就职公司等；所述接触属性可以包括如下信息：用户在某集团业务子公司所购买产品的产品属性，如用户在某集团业务子公司所购买的产险特定产品及所购买特定产品的数量等。
优选地，在本实施例中，所述进线流水特征包括，但不限于，用户通过特定进线方式(如语音电话进线方式)触发特定业务或产品，如电话咨询特定业务、购买特定产品、进行车险报案等。其中，所述进线流水特征可以根据时间顺序(如年/月/周/日)进行分类，例如，用户最近一年内所购买的特定产品，当日内的车险报案次数等。
所述获取模块202，用于通过预定分类算法(如Spark与Softmax结合的 多分类算法)，针对特定用户的用户特征和进线流水特征(如11000维)进行预定数量(如600类)的分类建模得到各分类模型，并获取各分类模型返回的各分类概率。
优选地，在本实施例中，每个分类节点通过不同的特征重要性进行建模。例如，在预测该特定用户是否进行车险报案模型中，重要特征包括：该特定用户是否购买车险、驾龄、及当日内的语音电话进线报案数据等，根据上述重要特征建模并判断该特定用户进行车险报案的概率，从而有效提升预测的准确性。
优选地，在本实施例中，所述每个分类节点通过不同的特征重要性进行建模包括：针对所有预定数量(600类)的分类，按照每个分类的特征重要性由高至低进行排序，并行选取每个分类中的特定特征(如前三位的特征)作为重要特征，根据每个分类中选取的重要特征进行建模。由于在建模的同时进行特征重要性分析，从而有效提升了业务预测的效率和精准度。
所述预测模块203，用于依据各分类模型返回的各分类概率进行排序，选取概率最高的分类预测该特定用户的业务意图。例如，返回概率最高的分类为产险业务，则预测该特定用户有购买产险的意图。
进一步地，在其它实施例中，所述预测模块203还用于：
通过预定时间内(如24小时内)该特定用户的反复进线率，获取该特定用户的优选推荐信息，从而实现精准推荐。例如，如果24小时内进行过车险报案的用户二次进线，则判定该用户有较大概率继续报案，因此，推荐车险报案的预测结果作为该用户的优选推荐信息。
进一步地，在其它实施例中，所述预测模块203还用于：
通过预设的语音识别算法(如DTW算法)，将预测结果(即预测的该特 定用户的业务意图)的文本内容转换为语音内容，并将该语音内容播报至该特定用户对应的用户端(如手持终端)。例如，所述语音内容可以设置为：“您是否需要办理xx业务(如产险业务)？”。
通过上述程序模块201-203，本申请所提出的用户意图预测系统20，通过结合语音识别技术与大数据处理技术，提高了用户意图预测的效率和准确度，将本申请应用于企业客服业务(如客户电话业务等)，可以大大缩短用户通过客服电话办理业务的时间，提高用户的满意度。
此外，本申请还提出一种用户意图预测方法。
参阅图3所示，是本申请用户意图预测方法一实施例的实施流程示意图。在本实施例中，根据不同的需求，图3所示的流程图中的步骤的执行顺序可以改变，某些步骤可以省略。
步骤S31，建立用户特征和进线流水特征。
优选地，在本实施例中，所述用户特征包括，但不限于，自然属性、集团属性、社会属性、接触属性(业务子公司)等。其中，所述自然属性可以包括如下信息：年龄、地区、性别、学历等；所述集团属性可以包括如下信息：用户在某集团所购买产品的产品属性，如产险用户等；所述社会属性可以包括如下信息：职业、子女、就职公司等；所述接触属性可以包括如下信息：用户在某集团业务子公司所购买产品的产品属性，如用户在某集团业务子公司所购买的产险特定产品及所购买特定产品的数量等。
优选地，在本实施例中，所述进线流水特征包括，但不限于，用户通过特定进线方式(如语音电话进线方式)触发特定业务或产品，如电话咨询特定业务、购买特定产品、进行车险报案等。其中，所述进线流水特征可以根据时间顺序(如年/月/周/日)进行分类，例如，用户最近一年内所购买的特定产品，当日内的车险报案次数等。
步骤S32，通过预定分类算法(如Spark与Softmax结合的多分类算法)，针对特定用户的用户特征和进线流水特征(如11000维)进行预定数量(如600类)的分类建模得到各分类模型，并获取各分类模型返回的各分类概率。
优选地，在本实施例中，每个分类节点通过不同的特征重要性进行建模。例如，在预测该特定用户是否进行车险报案模型中，重要特征包括：该特定用户是否购买车险、驾龄、及当日内的语音电话进线报案数据等，根据上述重要特征建模并判断该特定用户进行车险报案的概率，从而有效提升预测的准确性。
优选地，在本实施例中，所述每个分类节点通过不同的特征重要性进行建模包括：针对所有预定数量(600类)的分类，按照每个分类的特征重要性由高至低进行排序，并行选取每个分类中的特定特征(如前三位的特征)作为重要特征，根据每个分类中选取的重要特征进行建模。由于在建模的同时进行特征重要性分析，从而有效提升了业务预测的效率和精准度。
步骤S33，依据各分类模型返回的各分类概率进行排序，选取概率最高的分类预测该特定用户的业务意图。例如，返回概率最高的分类为产险业务，则预测该特定用户有购买产险的意图。
进一步地，在其它实施例中，所述用户意图预测方法还包括如下步骤：
通过预定时间内(如24小时内)该特定用户的反复进线率，获取该特定用户的优选推荐信息，从而实现精准推荐。例如，如果24小时内进行过车险报案的用户二次进线，则判定该用户有较大概率继续报案，因此，推荐车险报案的预测结果作为该用户的优选推荐信息。
进一步地，在其它实施例中，所述用户意图预测方法还包括如下步骤：
通过预设的语音识别算法(如DTW算法)，将预测结果(即预测的该特定用户的业务意图)的文本内容转换为语音内容，并将该语音内容播报至该特定用户对应的用户端(如手持终端)。例如，所述语音内容可以设置为：“您是否需要办理xx业务(如产险业务)？”。
通过上述步骤S31-S33及其它相关步骤，本申请所提出的用户意图预测方法，通过结合语音识别技术与大数据处理技术，提高了用户意图预测的效率和准确度，将本申请应用于企业客服业务(如客户电话业务等)，可以大大缩短用户通过客服电话办理业务的时间，提高用户的满意度。
进一步地，为实现上述目的，本申请还提供一种计算机可读存储介质(如ROM/RAM、磁碟、光盘)，所述计算机可读存储介质存储有用户意图预测系统20，所述用户意图预测系统20可被至少一个处理器22执行，以使所述至少一个处理器22执行如上所述的用户意图预测方法的步骤。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件来实现，但很多情况下前者是更佳的实施方式。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端设备(可以是手机，计算机，服务器，空调器，或者网络设备等)执行本申请各个实施例所述的方法。
以上参照附图说明了本申请的优选实施例，并非因此局限本申请的权利范围。上述本申请实施例序号仅仅为了描述，不代表实施例的优劣。另外，虽然在流程图中示出了逻辑顺序，但是在某些情况下，可以以不同于此处的顺序执行所示出或描述的步骤。
本领域技术人员不脱离本申请的范围和实质，可以有多种变型方案实现本申请，比如作为一个实施例的特征可用于另一实施例而得到又一实施例。凡是利用本申请说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本申请的专利保护范围内。

Claims (20)

  1. 一种电子设备，其特征在于，所述电子设备包括存储器、处理器，所述存储器上存储有可在所述处理器上运行的用户意图预测系统，所述用户意图预测系统被所述处理器执行时实现如下步骤：
    建立用户特征和进线流水特征；
    通过预定分类算法，针对特定用户的用户特征和进线流水特征进行预定数量的分类建模得到各分类模型，并获取各分类模型返回的各分类概率；及
    依据各分类模型返回的各分类概率进行排序，选取概率最高的分类预测该特定用户的业务意图。
  2. 如权利要求1所述的电子设备，其特征在于，所述用户特征包括自然属性、集团属性、社会属性、接触属性；及
    所述自然属性包括年龄、地区、性别、学历，所述集团属性包括用户在某集团所购买产品的产品属性，所述社会属性包括职业、子女、就职公司，所述接触属性包括用户在某集团业务子公司所购买产品的产品属性。
  3. 如权利要求1所述的电子设备，其特征在于，所述进线流水特征包括用户通过特定进线方式触发特定业务或产品，其中，所述进线流水特征根据时间顺序进行分类。
  4. 如权利要求1所述的电子设备，其特征在于，每个分类节点通过不同的特征重要性进行建模；
    所述每个分类节点通过不同的特征重要性进行建模包括：
    针对所有预定数量的分类，按照每个分类的特征重要性由高至低进行排序，并行选取每个分类中的特定特征作为重要特征，根据每个分类中选取的 重要特征进行建模。
  5. 如权利要求1所述的电子设备，其特征在于，所述用户意图预测系统被所述处理器执行时还用于实现如下步骤：
    通过预定时间内该特定用户的反复进线率，获取该特定用户的优选推荐信息；及
    通过预设的语音识别算法，将预测结果的文本内容转换为语音内容，并将该语音内容播报至该特定用户对应的用户端。
  6. 如权利要求2所述的电子设备，其特征在于，所述用户意图预测系统被所述处理器执行时还用于实现如下步骤：
    通过预定时间内该特定用户的反复进线率，获取该特定用户的优选推荐信息；及
    通过预设的语音识别算法，将预测结果的文本内容转换为语音内容，并将该语音内容播报至该特定用户对应的用户端。
  7. 如权利要求3所述的电子设备，其特征在于，所述用户意图预测系统被所述处理器执行时还用于实现如下步骤：
    通过预定时间内该特定用户的反复进线率，获取该特定用户的优选推荐信息；及
    通过预设的语音识别算法，将预测结果的文本内容转换为语音内容，并将该语音内容播报至该特定用户对应的用户端。
  8. 如权利要求4所述的电子设备，其特征在于，所述用户意图预测系统被所述处理器执行时还用于实现如下步骤：
    通过预定时间内该特定用户的反复进线率，获取该特定用户的优选推荐 信息；及
    通过预设的语音识别算法，将预测结果的文本内容转换为语音内容，并将该语音内容播报至该特定用户对应的用户端。
  9. 一种用户意图预测方法，应用于电子设备，其特征在于，所述方法包括：
    建立用户特征和进线流水特征；
    通过预定分类算法，针对特定用户的用户特征和进线流水特征进行预定数量的分类建模得到各分类模型，并获取各分类模型返回的各分类概率；及
    依据各分类模型返回的各分类概率进行排序，选取概率最高的分类预测该特定用户的业务意图。
  10. 如权利要求9所述的用户意图预测方法，其特征在于，所述用户特征包括自然属性、集团属性、社会属性、接触属性；
    所述自然属性包括年龄、地区、性别、学历，所述集团属性包括用户在某集团所购买产品的产品属性，所述社会属性包括职业、子女、就职公司，所述接触属性包括用户在某集团业务子公司所购买产品的产品属性。
  11. 如权利要求9所述的用户意图预测方法，其特征在于，所述进线流水特征包括用户通过特定进线方式触发特定业务或产品，其中，所述进线流水特征根据时间顺序进行分类。
  12. 如权利要求9所述的用户意图预测方法，其特征在于，每个分类节点通过不同的特征重要性进行建模；
    所述每个分类节点通过不同的特征重要性进行建模包括：
    针对所有预定数量的分类，按照每个分类的特征重要性由高至低进行排 序，并行选取每个分类中的特定特征作为重要特征，根据每个分类中选取的重要特征进行建模。
  13. 如权利要求9所述的用户意图预测方法，其特征在于，所述方法还包括步骤：
    通过预定时间内该特定用户的反复进线率，获取该特定用户的优选推荐信息；及
    通过预设的语音识别算法，将预测结果的文本内容转换为语音内容，并将该语音内容播报至该特定用户对应的用户端。
  14. 如权利要求10-12任一项所述的用户意图预测方法，其特征在于，所述方法还包括步骤：
    通过预定时间内该特定用户的反复进线率，获取该特定用户的优选推荐信息；及
    通过预设的语音识别算法，将预测结果的文本内容转换为语音内容，并将该语音内容播报至该特定用户对应的用户端。
  15. 一种计算机可读存储介质，所述计算机可读存储介质存储有用户意图预测系统，所述用户意图预测系统可被至少一个处理器执行，所述用户意图预测系统被所述处理器执行时实现如下步骤：
    建立用户特征和进线流水特征；
    通过预定分类算法，针对特定用户的用户特征和进线流水特征进行预定数量的分类建模得到各分类模型，并获取各分类模型返回的各分类概率；及
    依据各分类模型返回的各分类概率进行排序，选取概率最高的分类预测该特定用户的业务意图。
  16. 如权利要求15所述的计算机可读存储介质，其特征在于，所述用户特征包括自然属性、集团属性、社会属性、接触属性；及
    所述自然属性包括年龄、地区、性别、学历，所述集团属性包括用户在某集团所购买产品的产品属性，所述社会属性包括职业、子女、就职公司，所述接触属性包括用户在某集团业务子公司所购买产品的产品属性。
  17. 如权利要求15所述的计算机可读存储介质，其特征在于，所述进线流水特征包括用户通过特定进线方式触发特定业务或产品，其中，所述进线流水特征根据时间顺序进行分类。
  18. 如权利要求15所述的计算机可读存储介质，其特征在于，每个分类节点通过不同的特征重要性进行建模；
    所述每个分类节点通过不同的特征重要性进行建模包括：
    针对所有预定数量的分类，按照每个分类的特征重要性由高至低进行排序，并行选取每个分类中的特定特征作为重要特征，根据每个分类中选取的重要特征进行建模。
  19. 如权利要求15所述的计算机可读存储介质，其特征在于，所述用户意图预测系统被所述处理器执行时还用于实现如下步骤：
    通过预定时间内该特定用户的反复进线率，获取该特定用户的优选推荐信息；及
    通过预设的语音识别算法，将预测结果的文本内容转换为语音内容，并将该语音内容播报至该特定用户对应的用户端。
  20. 如权利要求16-18任一项所述的计算机可读存储介质，其特征在于，所述用户意图预测系统被所述处理器执行时还用于实现如下步骤：
    通过预定时间内该特定用户的反复进线率，获取该特定用户的优选推荐信息；及
    通过预设的语音识别算法，将预测结果的文本内容转换为语音内容，并将该语音内容播报至该特定用户对应的用户端。
PCT/CN2018/076177 2017-09-30 2018-02-10 用户意图预测方法、电子设备及计算机可读存储介质 WO2019061990A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710939717.6 2017-09-30
CN201710939717.6A CN107798341A (zh) 2017-09-30 2017-09-30 用户意图预测方法、电子设备及计算机可读存储介质

Publications (1)

Publication Number Publication Date
WO2019061990A1 true WO2019061990A1 (zh) 2019-04-04

Family

ID=61533200

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/076177 WO2019061990A1 (zh) 2017-09-30 2018-02-10 用户意图预测方法、电子设备及计算机可读存储介质

Country Status (2)

Country Link
CN (1) CN107798341A (zh)
WO (1) WO2019061990A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110427315A (zh) * 2019-06-18 2019-11-08 深圳壹账通智能科技有限公司 推优的测试装置、方法及存储介质
CN112529074A (zh) * 2020-12-09 2021-03-19 平安科技（深圳）有限公司 一种业务信息处理方法及相关设备
CN112835682A (zh) * 2021-02-25 2021-05-25 平安消费金融有限公司 一种数据处理方法、装置、计算机设备和可读存储介质
CN113822309A (zh) * 2020-09-25 2021-12-21 京东科技控股股份有限公司 用户的分类方法、装置和非易失性计算机可读存储介质

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108900342B (zh) * 2018-07-03 2021-06-11 阳光保险集团股份有限公司 电话业务处理方法、电话业务处理装置及服务器
CN109783608B (zh) * 2018-12-20 2021-01-05 出门问问信息科技有限公司 目标假设的确定方法、装置、可读存储介质和电子设备
CN111797874B (zh) * 2019-04-09 2024-04-09 Oppo广东移动通信有限公司 行为预测方法、装置、存储介质及电子设备
CN111091832B (zh) * 2019-11-28 2022-12-30 秒针信息技术有限公司 一种基于语音识别的意向评估方法和系统
CN113364920B (zh) * 2021-06-09 2023-01-20 中国银行股份有限公司 一种进线请求处理方法、装置及电子设备
CN113610504A (zh) * 2021-08-17 2021-11-05 平安资产管理有限责任公司 数据处理方法、装置、计算机设备及存储介质

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102238045A (zh) * 2010-04-27 2011-11-09 广州迈联计算机科技有限公司 一种无线互联网用户行为的预测系统和方法
CN103886487A (zh) * 2014-03-28 2014-06-25 焦点科技股份有限公司 基于分布式的b2b平台的个性化推荐方法与系统
US20150228275A1 (en) * 2014-02-10 2015-08-13 Mitsubishi Electric Research Laboratories, Inc. Statistical Voice Dialog System and Method
US20160004970A1 (en) * 2013-03-13 2016-01-07 Wei Lu Method and apparatus for recommendations with evolving user interests
CN106296242A (zh) * 2015-05-22 2017-01-04 苏宁云商集团股份有限公司 一种用于电子商务中商品推荐列表的生成方法及生成系统
CN106600369A (zh) * 2016-12-09 2017-04-26 广东奡风科技股份有限公司 基于朴素贝叶斯分类的银行金融产品实时推荐系统和方法
CN106993104A (zh) * 2016-11-14 2017-07-28 阿里巴巴集团控股有限公司 来电处理方法、装置及终端
CN107194515A (zh) * 2017-05-27 2017-09-22 冯小平 确定用户当前行为以及预测用户意图的方法和装置

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105721725A (zh) * 2016-02-03 2016-06-29 北京光年无限科技有限公司 一种面向客服的问答交互方法及系统

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102238045A (zh) * 2010-04-27 2011-11-09 广州迈联计算机科技有限公司 一种无线互联网用户行为的预测系统和方法
US20160004970A1 (en) * 2013-03-13 2016-01-07 Wei Lu Method and apparatus for recommendations with evolving user interests
US20150228275A1 (en) * 2014-02-10 2015-08-13 Mitsubishi Electric Research Laboratories, Inc. Statistical Voice Dialog System and Method
CN103886487A (zh) * 2014-03-28 2014-06-25 焦点科技股份有限公司 基于分布式的b2b平台的个性化推荐方法与系统
CN106296242A (zh) * 2015-05-22 2017-01-04 苏宁云商集团股份有限公司 一种用于电子商务中商品推荐列表的生成方法及生成系统
CN106993104A (zh) * 2016-11-14 2017-07-28 阿里巴巴集团控股有限公司 来电处理方法、装置及终端
CN106600369A (zh) * 2016-12-09 2017-04-26 广东奡风科技股份有限公司 基于朴素贝叶斯分类的银行金融产品实时推荐系统和方法
CN107194515A (zh) * 2017-05-27 2017-09-22 冯小平 确定用户当前行为以及预测用户意图的方法和装置

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110427315A (zh) * 2019-06-18 2019-11-08 深圳壹账通智能科技有限公司 推优的测试装置、方法及存储介质
CN113822309A (zh) * 2020-09-25 2021-12-21 京东科技控股股份有限公司 用户的分类方法、装置和非易失性计算机可读存储介质
CN112529074A (zh) * 2020-12-09 2021-03-19 平安科技（深圳）有限公司 一种业务信息处理方法及相关设备
CN112529074B (zh) * 2020-12-09 2024-04-26 平安科技（深圳）有限公司 一种业务信息处理方法及相关设备
CN112835682A (zh) * 2021-02-25 2021-05-25 平安消费金融有限公司 一种数据处理方法、装置、计算机设备和可读存储介质
CN112835682B (zh) * 2021-02-25 2024-04-05 平安消费金融有限公司 一种数据处理方法、装置、计算机设备和可读存储介质

Also Published As

Publication number Publication date
CN107798341A (zh) 2018-03-13

Similar Documents

Publication Publication Date Title
WO2019061990A1 (zh) 用户意图预测方法、电子设备及计算机可读存储介质
US11508012B2 (en) System and method for generating proactive underwriting offers using social media data
US20210224832A1 (en) Method and apparatus for predicting customer purchase intention, electronic device and medium
WO2019061991A1 (zh) 多元通用模型平台建模方法、电子设备及计算机可读存储介质
US10061822B2 (en) System and method for discovering and exploring concepts and root causes of events
CN111797210A (zh) 基于用户画像的信息推荐方法、装置、设备及存储介质
CN109067643B (zh) 基于关键字的回复方法、装置、计算机设备及存储介质
WO2019019636A1 (zh) 用户身份识别方法、电子装置及计算机可读存储介质
WO2017133615A1 (zh) 一种业务参数获取方法及装置
WO2019157779A1 (zh) 一种来电处理方法、电子装置及计算机可读存储介质
US20170270419A1 (en) Escalation prediction based on timed state machines
WO2019041522A1 (zh) 电子装置、保险推荐方法、及计算机可读存储介质
CN111124917B (zh) 公共测试用例的管控方法、装置、设备及存储介质
CN110349039B (zh) 投诉风险评估方法、系统、计算机设备及可读存储介质
US20220076140A1 (en) Prioritization of electronic communications
US20210110415A1 (en) Enhanced survey information synthesis
WO2019179030A1 (zh) 产品购买预测方法、服务器及存储介质
WO2019174184A1 (zh) 电话外呼分析控制方法、电子装置及可读存储介质
CN110162518B (zh) 数据分组方法、装置、电子设备及存储介质
WO2019071897A1 (zh) 实时推荐方法、电子设备及计算机可读存储介质
US20190095828A1 (en) Automatic ramp-up of controlled experiments
CN112016793B (zh) 基于目标用户群的资源分配方法、装置及电子设备
CN110807050B (zh) 性能分析方法、装置、计算机设备及存储介质
CN108536356A (zh) 坐席信息处理方法、装置及计算机可读存储介质
CN110827155A (zh) 信息处理方法、装置、电子设备及存储介质

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: 18861504

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 23/09/2020)

122 Ep: pct application non-entry in european phase

Ref document number: 18861504

Country of ref document: EP

Kind code of ref document: A1