WO2019227629A1 - 文本信息的生成方法、装置、计算机设备及存储介质 - Google Patents

文本信息的生成方法、装置、计算机设备及存储介质 Download PDF

Info

Publication number
WO2019227629A1
WO2019227629A1 PCT/CN2018/096329 CN2018096329W WO2019227629A1 WO 2019227629 A1 WO2019227629 A1 WO 2019227629A1 CN 2018096329 W CN2018096329 W CN 2018096329W WO 2019227629 A1 WO2019227629 A1 WO 2019227629A1
Authority
WO
WIPO (PCT)
Prior art keywords
user
personality
text information
characteristic words
words
Prior art date
Application number
PCT/CN2018/096329
Other languages
English (en)
French (fr)
Inventor
王杰
顾海倩
王姿雯
庄伯金
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 WO2019227629A1 publication Critical patent/WO2019227629A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/20Natural language analysis
    • G06F40/205Parsing
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/20Natural language analysis
    • G06F40/279Recognition of textual entities
    • G06F40/289Phrasal analysis, e.g. finite state techniques or chunking

Definitions

  • the present application relates to the field of computer technology, and in particular, to a method, an apparatus, a computer device, and a storage medium for generating text information.
  • the purpose of the present application is to provide a method, a device, a computer device, and a storage medium for generating text information, so as to generate text information that better reflects the personality characteristics of users.
  • the present application provides a method for generating text information, which includes the following contents: acquiring social text information of a user; acquiring character characteristic words of the user and word frequency of the character characteristic words according to the social text information; Input the personality characteristic words of the user and the word frequency of the personality characteristic words into a preset neural network-based predictive analysis system to obtain the personality characteristic vector of the user; obtain the initial characteristic words; convert the personality of the user The feature vector and the initial feature words are input to a preset personality generation model to generate personality feature text information matching the user.
  • the foregoing method further includes: obtaining an initial feature word.
  • the foregoing input of the user's personality feature vector into a preset personality generation model to generate the personality feature text information matching the user specifically includes:
  • the user's personality feature vector and initial feature words are input into a preset personality generation model to generate personality feature text information that matches the user.
  • the social text information includes at least one of a social message of the user and identity information of the user.
  • the obtaining the character characteristic words of the user and the word frequency of the character characteristic words according to the social text information includes: using word segmentation technology to perform word segmentation processing on the social text information to obtain the word segmentation processed phrase; according to The phrase after the word segmentation processing is matched with a preset psychological lexicon to obtain the personality characteristic words of the user and the word frequency of the personality characteristic words.
  • inputting the personality characteristic words of the user and the word frequency of the personality characteristic words into a preset neural network-based predictive analysis system to obtain a personality characteristic vector of the user include:
  • the user's personality characteristic words and the word frequency of the personality characteristic words are input into a preset neural network-based predictive analysis system to obtain the user's personality analysis value;
  • a user's personality feature vector is obtained according to the user's personality analysis value.
  • the personalization analysis value is a value between 0-1 or between 0-100.
  • the personalized text generation model is a memory sequence model.
  • the memory sequence model includes multiple memory units, and the multiple memory units are used to output multiple text words. The previous one of the multiple memory units The text word output by the memory unit is the input of the next memory unit.
  • the above-mentioned personality characteristic text information matched with the user includes the user's personality analysis information.
  • the above-mentioned personality feature text information matched with the user further includes chat strategy information matched with the user's personality analysis information.
  • the present application further provides a device for generating text information, and the device specifically includes:
  • the obtaining module is further configured to obtain the personality characteristic words of the user and the word frequency of the personality characteristic words according to the social text information;
  • a generating module for inputting the personality characteristic words of the user and the word frequency of the personality characteristic words into a preset neural network-based predictive analysis system to obtain the personality characteristic text information matching the user;
  • the above obtaining module is also used to obtain initial feature words
  • the generating module is further configured to input the personality feature vector and initial feature words of the user into a preset personality generation model, and generate personality feature text information matching the user.
  • the present application further provides a computer device including a memory, a processor, and a computer program stored on the memory and executable on the processor.
  • the processor is configured to:
  • the personality feature vector of the user is input to a preset personality generation model to generate personality feature text information that matches the user.
  • the present application also provides a computer-readable storage medium on which a computer program is stored.
  • the processor is configured to:
  • the personality feature vector of the user is input to a preset personality generation model to generate personality feature text information that matches the user.
  • the generation information of the text information includes:
  • a text information acquisition module for acquiring social text information of a user
  • a determining module configured to input personality characteristic words and / or personality-related text information of a user into a predictive analysis system based on a neural network technology for calculation and analysis to determine the personality characteristics of the user;
  • the personality characteristics of the user may be a personality characteristic vector of the user
  • a generation module configured to take the personality characteristics determined in the determination module as input, construct a personalized text generation, and use this to generate user personalized text information
  • An output module is used to output the user personality text information generated by the generating module.
  • the above-mentioned social text information of the user often includes a large number of keywords of the user's personality characteristics, and the user's personality characteristic keywords can be determined by the user's social text information and / or the user's historical conversation record information. For example, based on the user's social text information for a certain period of time, the user's personality characteristic keywords are obtained after analysis.
  • the user personalization text information may specifically be text information that matches the personality characteristics of the user.
  • the method, device, system, computer equipment, and storage medium for generating text information provided in the present application, obtain the personality characteristic words of the user and the word frequency of the personality characteristic words according to the social text information, and combine the preset predictive analysis system and personality generation
  • the model can generate text information that better reflects the personality characteristics of users.
  • FIG. 1 is a schematic flowchart of a method for generating text information according to an embodiment of the present application
  • FIG. 2 is a schematic diagram of a memory sequence generation model according to an embodiment of the present application.
  • FIG. 3 is a schematic block diagram of a device for generating text information according to an embodiment of the present application.
  • FIG. 4 is a schematic block diagram of a computer device according to an embodiment of the present application.
  • FIG. 5 is a schematic block diagram of a text information generating system according to an embodiment of the present application.
  • the method, device, computer equipment, and storage medium for generating text information provided in this application are applicable to insurance, banking, and other fields that need to identify the personality characteristics of users.
  • This application obtains a user's social text information, and according to the social text information, a character feature word and a word frequency that can describe the personality characteristics of the user.
  • a neural network-based predictive analysis system and a personality model are used to generate a user
  • the feature words and word frequencies of the feature words are analyzed to obtain text information that matches the personality characteristics of the user and better reflects the personality characteristics of the user.
  • FIG. 1 illustrates a method for generating text information according to an embodiment of the present application.
  • the method shown in FIG. 1 may be executed by a computer device, or may be executed by another device having a computing function.
  • the method shown in FIG. 1 includes steps 101 to 104. Steps 101 to 104 are described in detail below.
  • the social text information may include at least one of a user's social message and user's identity information.
  • the social message may be a social account registered by the user, a message posted by the user on social media, information forwarded or followed, friend information, chat information or chat information, and the like.
  • the user's identity information includes the user's ID number, mobile phone number, email address, age, gender, work and education.
  • the identity information of the user may be text information in a specific network database, for example, data accumulated by a company on a network data (cloud storage, cloud data) platform after a long period of time.
  • a web crawler technology can be used to grab the user's text information from the text information on the Internet or the text information on the network platform.
  • the big data analysis method is used to obtain the user's text information.
  • the specific big data analysis methods can include spark technology, Hadoop technology, and so on.
  • the user's personality characteristic words can also be referred to as the user's psychological characteristic words or key psychological characteristic words.
  • the personality characteristic words can reflect a user's personality characteristics or types, and after further analysis, it can be obtained that they can be better or better. Textual information that fully reflects the characteristics of the user.
  • acquiring the personality characteristic words of the user and the word frequency of the personality characteristic words according to the social text information specifically includes:
  • Word segmentation is performed on the user's text information by word segmentation technology to obtain the phrase after word segmentation processing;
  • the obtained social text information is "The sun is so good today, I am very happy, and I am very excited when I am happy.”
  • the user's personality characteristics can be obtained.
  • the words are happiness and excitement.
  • the frequency of happy words is 2 and the frequency of excited words is 1.
  • the aforementioned neural network-based predictive analysis system may be obtained through training in advance.
  • the neural network-based predictive analysis system can be trained through a large amount of historical data (the historical data may be the personality characteristic words of the user and the word frequency of the personality characteristic words, and the corresponding personality characteristic vector), so that the personality characteristics of the user Vectors can accurately and truly reflect the personality characteristics of users.
  • the user's personality characteristic words and the word frequency of the personality characteristic words are input into a preset neural network-based predictive analysis system to obtain the user's personality characteristic vector, which specifically includes:
  • the user's personality characteristic words and the word frequency of the personality characteristic words are input into a preset neural network-based predictive analysis system to obtain the user's personality analysis value;
  • a user's personality feature vector is obtained according to the user's personality analysis value.
  • the user's personality characteristic words can be used to determine the user's various personality analysis values in the five major personality classifications of psychology.
  • the value can be a value between 0-100 or 0. Values between -1.
  • the convolutional neural network can be used to analyze the personality characteristics of the user and the frequency of words according to the psychological personality classification to determine the analysis value of the user in each personality classification.
  • the personality feature vectors [a1, a2, a3, a4, a5] can be obtained based on the user's analysis values in each personality analysis classification.
  • an initial feature word may also be obtained.
  • the user's personality feature vector and the initial feature words can be input into a preset personality generation model to obtain the personality feature text information that matches the user.
  • the personality information text information matching the user may specifically include a personality analysis of the user.
  • the text information may further include information such as a chat strategy for such personality characteristics.
  • the above personified text generation model is a memory sequence model (also referred to as a memory sequence generation model).
  • the memory sequence model includes multiple memory units, and multiple memory units are used to output multiple text words.
  • the text word output by the previous memory unit is the input of the next memory unit.
  • a user's social text information is obtained, and according to the social text information, a personality feature word and a word frequency that can describe the personality characteristics of the user are obtained.
  • a neural network-based predictive analysis system and a personification generation model are used.
  • the user's feature words and the word frequency of the feature words are analyzed to obtain text information that matches the user's personality characteristics and can better reflect the user's personality characteristics.
  • FIG. 2 shows a schematic diagram of a memory sequence generation model according to an embodiment of the present application.
  • the and sequence generation model is composed of N memory subunits such as a first memory subunit, a second memory subunit, a third memory subunit, and an Nth memory subunit, where N is an integer greater than 1.
  • the output of the first memory sub-unit is used as the input of the second sub-unit, that is, in the memory sequence generation model shown in FIG. 2, the output of the i-th (i is an integer greater than 0 and less than or equal to N) memory sub-unit As input to the i + 1th memory subunit.
  • the personality feature vector [0,1,0,1,0] and the initial feature word “I” are used as model inputs, and after calculation of the first memory subunit, “Today” is output; then Next, use the personality feature vector [0,1,0,1,0] and the output of the first memory subunit as "today” as the input of the second memory subunit to output "mood”.
  • the method uses the personality feature vector [0,1,0,1,0] and the output "very” of the second memory subunit as the input of the third memory subunit, and outputs "very", and so on, until it passes through Figure 2.
  • the memory sequence generation model shown above obtains a complete text message.
  • a personified text generation model can be trained through a large amount of historical data (the historical data can be a user's personality feature vector and text information corresponding to the personality feature vector), so that the resulting text information can be more accurately and truthfully Reflect the personality characteristics of the user.
  • the method for generating text information in the embodiment of the present application is described in detail above with reference to FIGS. 1 and 2.
  • the device for generating text information in the embodiment of the present application is described below with reference to FIG. 3. It should be understood that the device in FIG. 3 can
  • Each step of the method for generating text information in the embodiment of the present application mentioned in FIG. 1 and FIG. 2 is performed, that is, the apparatus shown in FIG. 3 includes
  • each module omits duplication when describing the apparatus for generating text information in the embodiment of the present application as shown in FIG. 3.
  • FIG. 3 is a schematic block diagram of a device for generating text information according to an embodiment of the present application.
  • the apparatus shown in FIG. 3 includes:
  • An acquisition module 201 (also referred to as a text information acquisition module), configured to acquire social text information of a user;
  • the obtaining module 201 is further configured to obtain the personality characteristic words of the user and the word frequency of the personality characteristic words according to the social text information;
  • a generating module 202 is configured to input a personality characteristic word of a user and a word frequency of the personality characteristic word into a preset neural network-based predictive analysis system to obtain a personality characteristic vector of the user;
  • the generating module 202 is further configured to input the personality feature vector and initial feature words of the user into a preset personality generation model to generate personality feature text information matching the user.
  • the apparatus for generating text information can obtain a personality characteristic word of a user and a word frequency of the personality characteristic word according to social text information, and combined with a preset predictive analysis system and a personality generation model, can generate a user personality better Characteristic text messages.
  • the device 200 shown in FIG. 3 may specifically be a computer device or other device having a computing function.
  • Each module in the device 200 may be implemented by hardware or software, for example, a module in the device 200 It may be implemented by a hardware circuit, a field programmable gate array, etc., or may be implemented based on software, or part of the modules in the device 200 may be implemented by a hardware circuit and part of the modules may be implemented by software.
  • Embodiments of the present application also provide a computer device, such as a smart phone, tablet computer, notebook computer, desktop computer, rack server, blade server, tower server, or rack server (including stand-alone servers) that can execute programs. Or a server cluster consisting of multiple servers).
  • a computer device such as a smart phone, tablet computer, notebook computer, desktop computer, rack server, blade server, tower server, or rack server (including stand-alone servers) that can execute programs. Or a server cluster consisting of multiple servers).
  • the computer device 300 in the embodiment of the present application includes, but is not limited to, a memory 301 and a processor 302 that can be communicatively connected to each other through a system bus. It should be noted that FIG. 4 only shows a computer device 300 having components 301-230, but it should be understood that it is not required to implement all the illustrated components, and more or fewer components may be implemented instead.
  • the computer device 300 shown in FIG. 4 may also execute the method for generating text information in the embodiment of the present application, and the processor 302 in the computer device 300 may be equivalent to the obtaining module 201 and The generating module 202 and the processor 302 can implement functions performed by three modules in the text information generating device 200.
  • the memory 301 (readable storage medium) includes a flash memory, a hard disk, a multimedia card, a card-type memory (for example, 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 disks, optical disks, etc.
  • the memory 301 may be an internal storage unit of the computer device 300, such as a hard disk or a memory of the computer device 300.
  • the memory 301 may also be an external storage device of the computer device 300, such as a plug-in hard disk, a smart memory card (SMC), and a secure digital (Secure Digital, SD) card, Flash card, etc.
  • the memory 301 may also include both the internal storage unit of the computer device 300 and its external storage device.
  • the memory 301 is generally used to store an operating system and various types of application software installed in the computer device 300, such as program codes of the text information generating device 200 shown in FIG.
  • the memory 301 can also be used to temporarily store various types of data that have been output or will be output.
  • the processor 302 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip in some embodiments.
  • the processor 302 is generally used to control the overall operation of the computer device 300.
  • the processor 302 is configured to run program code or process data stored in the memory 301, for example, to run the agent task management apparatus 10 to implement the method for generating text information in the first embodiment.
  • FIG. 5 is a schematic block diagram of a text information generating system according to an embodiment of the present application.
  • the text information generation system shown in FIG. 5 may be composed of the text information generation device 200 or the computer device 300 described above, and the text system may also perform each step in the text information generation method in the embodiment of the present application.
  • the text information generating system shown in FIG. 5 is described below.
  • the text information generation system 400 includes a text information acquisition module 401, a determination module 402, a generation module 403, and an output module 404.
  • the text information acquisition module 401 is configured to acquire social text information of a user.
  • the user's social text information often contains a large number of keywords of the user's personality characteristics, and the user's personality characteristic keywords can be determined by the user's social text information and / or the user's historical conversation record information. For example, based on the user's social text information for a certain period of time, the user's personality characteristic keywords are obtained after analysis.
  • the determining module 402 is configured to input personality characteristic words and / or personality-related text information of a user into a predictive analysis system based on a neural network technology for calculation and analysis to determine the personality characteristics of the user.
  • the personality characteristic of the user may be a personality characteristic vector of the user.
  • the generating module 403 is configured to take the personality characteristics determined in the determining module as input, construct a personalized text generation, and use this to generate user personalized text information.
  • the user personalization text information may specifically be text information that matches the personality characteristics of the user.
  • the output module 404 is configured to output the user personality text information generated by the generating module.
  • the text information acquisition module 401 corresponds to the acquisition module 201 in the text information generation device 200
  • the determination module 402 corresponds to the acquisition module 201 in the text information generation device 200
  • the generation module 403 corresponds to the text information generation device 200.
  • Acquisition module 201 Compared with the text information generating device 200, the text information generating system 400 has an output module 404.
  • the output module 404 can output the user's personality text information for further analysis or use.
  • the text information acquisition module 401, the determination module 402, and the generation module 403 in the text information generation system 400 are equivalent to the processor 302 in the computer device 300 and are used to generate user-personalized text information.
  • the text information generating system 400 may be composed of one device, or may be composed of multiple devices.
  • This embodiment also provides a computer-readable storage medium, such as a flash memory, a hard disk, a multimedia card, a card-type memory (for example, SD or DX memory, etc.), a random access memory (RAM), a static random access memory (SRAM), Read memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disks, optical disks, servers, App application stores, etc., which have computer programs stored on them, When the program is executed by the processor, the corresponding function is realized.
  • the computer-readable storage medium of this embodiment is used to store the agent task management apparatus 10, and when executed by a processor, implements the method for generating text information of the first embodiment.
  • the disclosed systems, devices, and methods may be implemented in other ways.
  • the device embodiments described above are only schematic.
  • the division of the unit is only a logical function division.
  • multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, which may be electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objective of the solution of this embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each of the units may exist separately physically, or two or more units may be integrated into one unit.
  • the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.
  • the technical solution of the present application is essentially a part that contributes to the existing technology or a part of the technical solution can be embodied in the form of a software product.
  • the computer software product is stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in the embodiments of the present application.
  • the foregoing storage media include: U disks, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks or compact discs and other media that can store program codes .

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Machine Translation (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

提供一种文本信息的生成方法、装置、计算机设备及存储介质。该方法包括:获取用户的社交文本信息(101);从社交文本信息中提取用户的性格特征词及词频(102);将用户的性格特征词以及词频输入到预先设置的预测分析系统,得到用户的人格特征向量(103);将用户的人格特征向量输入到预先设置的人格化生成模型,生成与用户人格特征匹配的文本信息(104)。该方法能够生成更好地体现用户人格特点的文本信息。

Description

文本信息的生成方法、装置、计算机设备及存储介质
本申请申明享有2018年5月30日递交的申请号为2018105379900、名称为“文本信息的生成方法、装置、计算机设备及存储介质”的中国专利申请的优先权,该中国专利申请的整体内容以参考的方式结合在本申请中。
技术领域
本申请涉及计算机技术领域,尤其涉及一种文本信息的生成方法、装置、计算机设备及存储介质。
背景技术
在很多场合下,例如,在银行和保险行业,为了了解用户的特点并与用户进行更好的交谈和互动,目前在一些系统中出现了根据用户的身份、年龄和国家等信息生成的用户的文本信息。但是这些文本信息反映的用户的信息有限,不能较好地描述用户的特点。
发明内容
本申请的目的是提供一种文本信息的生成方法、装置、计算机设备及存储介质,用于生成更好地体现用户人格特点的文本信息。
为实现上述目的,本申请提供一种文本信息的生成方法,包括以下内容:获取用户的社交文本信息;根据所述社交文本信息获取所述用户的性格特征词以及所述性格特征词的词频;将所述用户的性格特征词以及所述性格特征词的词频输入到预先设置的基于神经网络的预测分析系统,得到所述用户的人格特征向量;获取初始特征字词;将所述用户的人格特征向量和所述初始特征字词输入到预先设置的人格化生成模型,生成与所述用户匹配的人格特 征文本信息。
在一种可能的实现方式中,上述方法还包括:获取初始特征字词。
上述将用户的人格特征向量输入到预先设置的人格化生成模型,生成与用户匹配的人格特征文本信息,具体包括:
将用户的人格特征向量和初始特征字词输入到预先设置的人格化生成模型,生成与用户匹配的人格特征文本信息。
在一种可能的实现方式中,上述社交文本信息包括用户的社交消息和用户的身份信息中的至少一种。
在一种可能的实现方式中,上述根据社交文本信息获取所述用户的性格特征词以及性格特征词的词频,包括:采用分词技术对社交文本信息进行分词处理,得到分词处理后的词组;根据分词处理后的词组以及预设的心理学词库进行匹配,得到所述用户的性格特征词以及性格特征词的词频。
在一种可能的实现方式中,将用户的性格特征词以及所述性格特征词的词频输入到预先设置的基于神经网络的预测分析系统,得到用户的人格特征向量,包括:
将用户的性格特征词以及性格特征词的词频输入到预先设置的基于神经网络的预测分析系统,得到用户的人格化分析值;
根据用户的人格化分析值得到用户的人格特征向量。
在一种可能的实现方式中,上述人格化分析值为0-1之间或者0-100之间的数值。
在一种可能的实现方式中,人格化的文本生成模型为一种记忆序列模型,记忆序列模型包括多个记忆单元,多个记忆单元用于输出多个文本词,多个记忆单元中的前一个记忆单元输出的文本词为下一个记忆单元的输入。
在一种可能的实现方式中,上述与用户匹配的人格特征文本信息包括用户的人格分析信息。
在一种可能的实现方式中,上述与用户匹配的人格特征文本信息还包括 与用户的人格分析信息相匹配的交谈策略信息。
为实现上述目的,本申请还提供一种文本信息的生成装置,该装置具体包括:
获取模块,用于获取用户的社交文本信息;
获取模块还用于根据社交文本信息获取用户的性格特征词以及性格特征词的词频;
生成模块,用于将用户的性格特征词以及性格特征词的词频输入到预先设置的基于神经网络的预测分析系统,得到与用户匹配的人格特征文本信息;
上述获取模块还用于获取初始特征字词;
生成模块还用于将用户的人格特征向量和初始特征字词输入到预先设置的人格化生成模型,生成与用户匹配的人格特征文本信息。
为实现上述目的,本申请还提供一种计算机设备,包括存储器、处理器以及存储在存储器上并可在处理器上运行的计算机程序,所述处理器用于:
获取用户的社交文本信息;
根据所述社交文本信息获取所述用户的性格特征词以及所述性格特征词的词频;
将所述用户的性格特征词以及所述性格特征词的词频输入到预先设置的基于神经网络的预测分析系统,得到所述用户的人格特征向量;
将所述用户的人格特征向量输入到预先设置的人格化生成模型,生成与所述用户匹配的人格特征文本信息。
为实现上述目的,本申请还提供计算机可读存储介质,其上存储有计算机程序,所述计算机程序被处理器执行时,所述处理器用于:
获取用户的社交文本信息;
根据所述社交文本信息获取所述用户的性格特征词以及所述性格特征词的词频;
将所述用户的性格特征词以及所述性格特征词的词频输入到预先设置的基于神经网络的预测分析系统,得到所述用户的人格特征向量;
将所述用户的人格特征向量输入到预先设置的人格化生成模型,生成与所述用户匹配的人格特征文本信息。
为实现上述目的,本申请还提出了一种文本信息的生成系统,该文本信息的生成信息包括:
文本信息获取模块,用于获取用户的社交文本信息;
确定模块,用于将用户的性格特征词和/或性格相关的文本信息输入基于神经网络技术的预测分析系统中进行计算分析,确定用户的人格特征;
上述用户的人格特征具体可以是用户的人格特征向量;
生成模块,用于将确定模块中确定的人格特征作为输入,构建人格化文本生成,并以此来生成用户人格化文本信息;
输出模块,用于将生成模块生成的用户人格文本信息进行输出。
上述用户社交文本信息往往包含大量用户性格特征的关键词,可以通过用户的社交文本信息和/或用户的历史交谈记录信息等来确定用户的性格特征关键词。例如,通过用户的某一段时间区间的社交文本信息,经过分析来获取用户性格特征关键词。
上述用户人格化文本信息具体可以是与用户的人格特征匹配的文本信息。
本申请提供的文本信息的生成方法、装置、系统、计算机设备及存储介质,通过根据社交文本信息获取用户的性格特征词以及该性格特征词的词频,并结合预先设置的预测分析系统和人格生成模型,能够生成更好地体现用户人格特点的文本信息。
附图说明
图1为本申请实施例的文本信息的生成方法的示意性流程图;
图2为本申请实施例的记忆序列生成模型的示意图;
图3为本申请实施例的文本信息的生成装置的示意性框图;
图4是本申请实施例的计算机设备的示意性框图;
图5是本申请实施例的文本信息的生成系统的示意性框图。
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。
本申请提供的文本信息的生成方法、装置、计算机设备及存储介质,适用于保险、银行以及其它需要识别用户人格特征的领域。本申请通过获取用户的社交文本信息,并根据该社交文本信息中获取能够描述该用户性格特征的性格特征词以及词频,接下来,再通过基于神经网络的预测分析系统和人格化生成模型,对用户的特征词以及特征词的词频进行分析,从而得到与用户人格特征相匹配,并且能够更好地反应用户人格特征的文本信息。
下面结合图1和图2对本申请实施例的文本信息的生成方法进行详细的描述。
图1示出了本申请实施例的文本信息的生成方法。图1所示的方法可以由计算机设备执行,或者由其它具有运算功能的设备来执行。图1所示的方法包括步骤101至步骤104,下面分别对步骤101至步骤104进行详细的描述和介绍。
101、获取用户的社交文本信息。
上述社交文本信息可以包括用户的社交消息和用户的身份信息中的至少 一种。
具体地,上述社交消息可以是用户注册的社交账号,用户在社交媒体上发布的消息,转发或者关注的信息,好友信息,聊天信息或者交谈信息等。
用户的身份信息包括用户的身份证号码、手机号、邮箱、年龄、性别、工作和学历等。
上述用户的身份信息具体可以是某一具体网络数据库中的文本信息,例如,某公司经过长时间积累在网络数据(云存储、云数据)平台上的数据。对于互联网中的文本信息,为了获取用户的文本信息,可以采用网络爬虫技术从互联网中的文本信息或者网络平台上的文本信息中抓取用户的文本信息,而对于网络数据平台上的数据,可以采用大数据分析的方法来获取用户的文本信息,具体的大数据分析方法可以包括spark技术、Hadoop技术等。
102、根据社交文本信息获取用户的性格特征词以及性格特征词的词频。
上述用户的性格特征词也可以称为用户的心理学特征词或者关键心理特征词,该性格特征词能够反映出一个用户的性格特点或者类型,并且经过进一步的分析之后可以得到能够较好或者较为全面地反映用户性格特征的文本信息。
可选地,根据社交文本信息获取用户的性格特征词以及性格特征词的词频,具体包括:
通过分词技术对用户的文本信息进行分词处理,得到分词处理后的词组;
将分词处理后的词组与预先建立的心理学词库进行匹配,确定用户文本信息中性格特征词以及该性格特征词出现的词频。
例如,获取的社交文本信息为“今天太阳太好了,我的心情非常高兴,在高兴之余还很兴奋”,通过将该社交文本信息与心理学词库进行匹配,可以得到用户的性格特征词为高兴和兴奋,其中,高兴的词频为2,兴奋的词频为1。
103、将用户的性格特征词以及性格特征词的词频输入到预先设置的基于 神经网络的预测分析系统,得到用户的人格特征向量。
应理解,上述基于神经网络的预测分析系统可以是预先通过训练得到的。具体地,可以通过大量的历史数据(该历史数据可以是用户的性格特征词以及性格特征词的词频,以及对应的人格特征向量)来训练该基于神经网络的预测分析系统,使得用户的人格特征向量能够较为准确和真实地反映用户的性格特征。
具体地,将用户的性格特征词以及性格特征词的词频输入到预先设置的基于神经网络的预测分析系统,得到用户的人格特征向量,具体包括:
将用户的性格特征词以及性格特征词的词频输入到预先设置的基于神经网络的预测分析系统,得到用户的人格化分析值;
根据用户的人格化分析值得到用户的人格特征向量。
在确定用户的人格化分析值时,具体可以根据用户的性格特征词确定用户在心理学五大人格分类中各种人格的分析值,该数值可以具体为0-100之间的数值,也可以为0-1之间的数值。
具体地,可以通过卷积神经网络将用户的人格特征词以及词频按照心理学人格分类进行匹配度分析,从而确定用户在各个人格分类中的分析值。例如,通过卷积神经网络对获取的用户心理特征词和词频进行分析,确定用户的开放性维度的值为a1=60,责任心维度的值为a2=50,外倾性维度的值为a3=70,宜人性维度的值为a4=80,神经质维度的值为a5=75。接下来,再根据用户在各个人格分析分类中的分析值就可以得到人格特征向量[a1,a2,a3,a4,a5]。
104、将用户的人格特征向量输入到预先设置的人格化生成模型,生成与用户匹配的人格特征文本信息。
可选地,在获取与用户匹配的人格特征文本信息之前,还可以获取初始特征字词。
在获取了初始特征字词之后,可以将用户的人格特征向量以及初始特征 字词输入到预先设置的人格化生成模型,从而得到与用户匹配的人格特征文本信息。
可选地,上述与用户匹配的人格特征文本信息具体可以包括用户的人格分析,进一步地,上述文本信息还可以包括针对这种人格特征的交谈策略等信息。
上述人格化的文本生成模型为一种记忆序列模型(也可以称为记忆序列生成模型),该记忆序列模型包括多个记忆单元,多个记忆单元用于输出多个文本词,多个记忆单元中的前一个记忆单元输出的文本词为下一个记忆单元的输入。
本申请中,通过获取用户的社交文本信息,并根据该社交文本信息获取能够描述该用户性格特征的性格特征词以及词频,接下来,再通过基于神经网络的预测分析系统和人格化生成模型,对用户的特征词以及特征词的词频进行分析,从而得到与用户人格特征相匹配,并且能够更好地反应用户人格特征的文本信息。
图2示出了本申请实施例的记忆序列生成模型的示意图。该以及序列生成模型由第一记忆子单元、第二记忆子单元、第三记忆子单元以及第N记忆子单元等N个记忆子单元组成,其中,N为大于1的整数。第一记忆子单元的输出作为第二子单元的输入,也就是说,在图2所示的记忆序列生成模型中,第i(i为大于0小于或者等于N的整数)记忆子单元的输出作为第i+1记忆子单元的输入。
具体地,如图2所示,人格特征向量[0,1,0,1,0]与初始特征字词“我”作为模型输入,经过第一记忆子单元的计算,输出“今天”;接下来,将人格特征向量[0,1,0,1,0]以及第一记忆子单元的输出的“今天”作为第二记忆子单元的输入,输出“心情”;接下来,再按照类似的方式将人格特征向量[0,1,0,1,0]以及第二记忆子单元的输出的“很”作为第三记忆子单元的输入,输出“很”,以此类推,直到通过图2所示的记忆序列生成模型得到一 段完整的文本信息为止。
应理解,上述人格化文本生成模型可以是预先通过训练得到的。
具体地,可以通过大量的历史数据(该历史数据可以是用户的人格特征向量以及与该人格特征向量对应的文本信息)来训练人格化文本生成模型,使得最终得到的文本信息能够较为准确和真实地反映用户的性格特征。
上文结合图1和图2对本申请实施例的文本信息的生成方法进行了详细的描述,下面结合图3对本申请实施例的文本信息的生成装置进行描述,应理解,图3中的装置能够执行上文中图1和图2中所涉及的本申请实施例的文本信息的生成方法的各个步骤,也就是说,图3所示的装置包括能够执行本申请实施例的文本信息的生成方法的各个模块,为了简洁,下面在对图3对本申请实施例的文本信息的生成装置进行描述时适当省略重复的内容。
图3为本申请实施例的文本信息的生成装置的示意性框图。图3所示的装置包括:
获取模块201(也可以称为文本信息获取模块),用于获取用户的社交文本信息;
获取模块201还用于根据社交文本信息获取用户的性格特征词以及性格特征词的词频;
生成模块202,用于将用户的性格特征词以及性格特征词的词频输入到预先设置的基于神经网络的预测分析系统,得到用户的人格特征向量;
生成模块202还用于将用户的人格特征向量和初始特征字词输入到预先设置的人格化生成模型,生成与用户匹配的人格特征文本信息。
本申请实施例的文本信息的生成装置通过根据社交文本信息获取用户的性格特征词以及该性格特征词的词频,并结合预先设置的预测分析系统和人格生成模型,能够生成更好地体现用户人格特点的文本信息。
应理解,图3所示的装置200具体可以是计算机设备,或者其它具有运算功能的设备,装置200中的各个模块可以由硬件来实现,也可以由软件来 实现,例如,装置200中的模块可以通过硬件电路,现场可编程门阵列等实现,也可以基于软件来实现,或者,装置200中的部分模块通过硬件电路实现,部分模块通过软件来实现。
本申请实施例还提供一种计算机设备,如可以执行程序的智能手机、平板电脑、笔记本电脑、台式计算机、机架式服务器、刀片式服务器、塔式服务器或机柜式服务器(包括独立的服务器,或者多个服务器所组成的服务器集群)等。
如图4所示,本申请实施例的计算机设备300至少包括但不限于:可通过系统总线相互通信连接的存储器301、处理器302。需要指出的是,图4仅示出了具有组件301-302的计算机设备300,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。
应理解,图4所示的计算机设备300也可以执行本申请实施例的文本信息的生成方法,并且,计算机设备300中的处理器302可以相当于文本信息的生成装置200中的获取模块201和生成模块202,处理器302能够实现文本信息的生成装置200中的三个模块执行的功能。
本实施例中,存储器301(即可读存储介质)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,存储器301可以是计算机设备300的内部存储单元,例如该计算机设备300的硬盘或内存。在另一些实施例中,存储器301也可以是计算机设备300的外部存储设备,例如该计算机设备300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,存储器301还可以既包括计算机设备300的内部存储单元也包括其外部存储设备。本实施例中,存储器301通常用于存储安装于计算机设备300的操作系统和各类应用软件,例如图2所示的文本 信息的生成装置200的程序代码等。此外,存储器301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。
处理器302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302通常用于控制计算机设备300的总体操作。本实施例中,处理器302用于运行存储器301中存储的程序代码或者处理数据,例如运行坐席任务管理装置10,以实现实施例一的文本信息的生成方法。
图5是本申请实施例的文本信息的生成系统的示意性框图。图5所示的文本信息的生成系统可以由上文中的文本信息的生成装置200或者计算机设备300组成,该文本系统也可以执行本申请实施例的文本信息的生成方法中的各个步骤。下面对图5所示的文本信息的生成系统进行描述。
如图5所示,文本信息的生成系统400包括:文本信息获取模块401,确定模块402、生成模块403和输出模块404。
其中,文本信息获取模块401用于获取用户的社交文本信息。
用户社交文本信息往往包含大量用户性格特征的关键词,可以通过用户的社交文本信息和/或用户的历史交谈记录信息等来确定用户的性格特征关键词。例如,通过用户的某一段时间区间的社交文本信息,经过分析来获取用户性格特征关键词。
确定模块402用于将用户的性格特征词和/或性格相关的文本信息输入基于神经网络技术的预测分析系统中进行计算分析,确定用户的人格特征。
上述用户的人格特征具体可以是用户的人格特征向量。
生成模块403用于将确定模块中确定的人格特征作为输入,构建人格化文本生成,并以此来生成用户人格化文本信息。
上述用户人格化文本信息具体可以是与用户的人格特征匹配的文本信息。
输出模块404用于将生成模块生成的用户人格文本信息进行输出。
其中,文本信息获取模块401相当于文本信息的生成装置200中的获取 模块201,确定模块402相当于文本信息的生成装置200中的获取模块201,生成模块403相当于文本信息的生成装置200中的获取模块201。与文本信息的生成装置200相比,该文本信息的生成系统400多出一个输出模块404,该输出模块404可以将用户的人格文本信息输出,便于进一步的分析或者使用。
文本信息的生成系统400中的文本信息获取模块401,确定模块402、生成模块403相当于计算机设备300中的处理器302,用于生成用户人格化文本信息。
另外,文本信息的生成系统400可以由一个设备组成,也可以由多个设备组成。
本实施例还提供一种计算机可读存储介质,如闪存、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、服务器、App应用商城等等,其上存储有计算机程序,程序被处理器执行时实现相应功能。本实施例的计算机可读存储介质用于存储坐席任务管理装置10,被处理器执行时实现实施例一的文本信息的生成方法。
本领域普通技术人员可以意识到,结合本文中所公开的实施例描述的各示例的单元及算法步骤,能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行,取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能,但是这种实现不应认为超出本申请的范围。
所属领域的技术人员可以清楚地了解到,为描述的方便和简洁,上述描述的系统、装置和单元的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。
在本申请所提供的实施例中,应该理解到,所揭露的系统、装置和方法,可以通过其它的方式实现。例如,以上所描述的装置实施例仅仅是示意性的, 例如,所述单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,例如多个单元或组件可以结合或者可以集成到另一个系统,或一些特征可以忽略,或不执行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口,装置或单元的间接耦合或通信连接,可以是电性,机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外,在本申请各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储介质中。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或者该技术方案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中,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括:U盘、移动硬盘、只读存储器(Read-Only Memory,ROM)、随机存取存储器(Random Access Memory,RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述,仅为本申请的具体实施方式,但本申请的保护范围并不局限于此,任何熟悉本技术领域的技术人员在本申请揭露的技术范围内,可轻易想到变化或替换,都应涵盖在本申请的保护范围之内。因此,本申请的保护范围应以所述权利要求的保护范围为准。

Claims (20)

  1. 一种文本信息的生成方法,其特征在于,包括:
    获取用户的社交文本信息;
    根据所述社交文本信息获取所述用户的性格特征词以及所述性格特征词的词频;
    将所述用户的性格特征词以及所述性格特征词的词频输入到预先设置的基于神经网络的预测分析系统,得到所述用户的人格特征向量;
    将所述用户的人格特征向量输入到预先设置的人格化生成模型,生成与所述用户匹配的人格特征文本信息。
  2. 根据权利要求1所述的方法,其特征在于,所述社交文本信息包括所述用户的社交消息和所述用户的身份信息中的至少一种。
  3. 根据权利要求1所述的方法,其特征在于,所述根据所述社交文本信息获取所述用户的性格特征词以及所述性格特征词的词频,包括:
    采用分词技术对所述社交文本信息进行分词处理,得到分词处理后的词组;
    根据所述分词处理后的词组以及预设的心理学词库进行匹配,得到所述用户的性格特征词以及所述性格特征词的词频。
  4. 根据权利要求1所述的方法,其特征在于,将所述用户的性格特征词以及所述性格特征词的词频输入到预先设置的基于神经网络的预测分析系统,得到所述用户的人格特征向量,包括:
    将所述用户的性格特征词以及所述性格特征词的词频输入到预先设置的基于神经网络的预测分析系统,得到所述用户的人格化分析值;
    根据所述用户的人格化分析值得到所述用户的人格特征向量。
  5. 根据权利要求4所述的方法,其特征在于,所述人格化分析值为0-1之间或者0-100之间的数值。
  6. 根据权利要求1至5中任一项所述的方法,其特征在于,所述人格化的文本生成模型为一种记忆序列模型,所述记忆序列模型包括多个记忆单元,所述多个记忆单元用于输出多个文本词,所述多个记忆单元中的前一个记忆单元输出的文本词为下一个记忆单元的输入。
  7. 根据权利要求1至5中任一项所述的方法,其特征在于,所述与所述用户人格特征匹配的文本信息包括所述用户的人格分析信息。
  8. 一种文本信息的生成装置,其特征在于,包括:
    获取模块,用于获取用户的社交文本信息;
    所述获取模块还用于根据所述社交文本信息获取所述用户的性格特征词以及所述性格特征词的词频;
    生成模块,用于将所述用户的性格特征词以及所述性格特征词的词频输入到预先设置的基于神经网络的预测分析系统,得到所述用户的人格特征向量;
    所述生成模块还用于将所述用户的人格特征向量和所述初始特征字词输入到预先设置的人格化生成模型,生成与所述用户匹配的人格特征文本信息。
  9. 一种计算机设备,包括存储器、处理器以及存储在存储器上并可在处理器上运行的计算机程序,所述处理器用于:
    获取用户的社交文本信息;
    根据所述社交文本信息获取所述用户的性格特征词以及所述性格特征词的词频;
    将所述用户的性格特征词以及所述性格特征词的词频输入到预先设置的基于神经网络的预测分析系统,得到所述用户的人格特征向量;
    将所述用户的人格特征向量输入到预先设置的人格化生成模型,生成与所述用户匹配的人格特征文本信息。
  10. 根据权利要求9所述的计算机设备,其特征在于,所述社交文本信息包括所述用户的社交消息和所述用户的身份信息中的至少一种。
  11. 根据权利要求9所述的计算机设备,其特征在于,所述处理器用于:
    采用分词技术对所述社交文本信息进行分词处理,得到分词处理后的词组;
    根据所述分词处理后的词组以及预设的心理学词库进行匹配,得到所述用户的性格特征词以及所述性格特征词的词频。
  12. 根据权利要求9所述的计算机设备,其特征在于,所述处理器用于:
    将所述用户的性格特征词以及所述性格特征词的词频输入到预先设置的基于神经网络的预测分析系统,得到所述用户的人格化分析值;
    根据所述用户的人格化分析值得到所述用户的人格特征向量。
  13. 根据权利要求12所述的计算机设备,其特征在于,所述人格化分析值为0-1之间或者0-100之间的数值。
  14. 根据权利要求9至13中任一项所述的计算机设备,其特征在于,所述人格化的文本生成模型为一种记忆序列模型,所述记忆序列模型包括多个记忆单元,所述多个记忆单元用于输出多个文本词,所述多个记忆单元中的前一个记忆单元输出的文本词为下一个记忆单元的输入。
  15. 根据权利要求9至13中任一项所述的计算机设备,其特征在于,所述与所述用户人格特征匹配的文本信息包括所述用户的人格分析信息。
  16. 一种计算机可读存储介质,其上存储有计算机程序,其特征在于:所述计算机程序被处理器执行时实现权利要求1至7任一项所述方法的步骤。
  17. 根据权利要求16所述的计算机可读存储介质,其特征在于,所述社 交文本信息包括所述用户的社交消息和所述用户的身份信息中的至少一种。
  18. 根据权利要求16所述的计算机可读存储介质,其特征在于,所述根据所述社交文本信息获取所述用户的性格特征词以及所述性格特征词的词频,包括:
    采用分词技术对所述社交文本信息进行分词处理,得到分词处理后的词组;
    根据所述分词处理后的词组以及预设的心理学词库进行匹配,得到所述用户的性格特征词以及所述性格特征词的词频。
  19. 根据权利要求16所述的计算机可读存储介质,其特征在于,将所述用户的性格特征词以及所述性格特征词的词频输入到预先设置的基于神经网络的预测分析系统,得到所述用户的人格特征向量,包括:
    将所述用户的性格特征词以及所述性格特征词的词频输入到预先设置的基于神经网络的预测分析系统,得到所述用户的人格化分析值;
    根据所述用户的人格化分析值得到所述用户的人格特征向量。
  20. 根据权利要求19所述的计算机可读存储介质,其特征在于,所述人格化分析值为0-1之间或者0-100之间的数值。
PCT/CN2018/096329 2018-05-30 2018-07-19 文本信息的生成方法、装置、计算机设备及存储介质 WO2019227629A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810537990.0 2018-05-30
CN201810537990.0A CN108829668B (zh) 2018-05-30 2018-05-30 文本信息的生成方法、装置、计算机设备及存储介质

Publications (1)

Publication Number Publication Date
WO2019227629A1 true WO2019227629A1 (zh) 2019-12-05

Family

ID=64146358

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/096329 WO2019227629A1 (zh) 2018-05-30 2018-07-19 文本信息的生成方法、装置、计算机设备及存储介质

Country Status (2)

Country Link
CN (1) CN108829668B (zh)
WO (1) WO2019227629A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111581923A (zh) * 2020-04-29 2020-08-25 北京字节跳动网络技术有限公司 文案生成方法、装置、设备和计算机可读存储介质
CN113837445A (zh) * 2021-08-27 2021-12-24 合肥工业大学 基于注意力机制的人格预测方法及系统

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113157941B (zh) * 2021-04-08 2023-04-18 支付宝(杭州)信息技术有限公司 业务特征数据处理、文本生成方法、装置及电子设备
CN115952272B (zh) * 2023-03-10 2023-05-26 杭州心识宇宙科技有限公司 一种生成对话信息的方法、装置、设备及可读存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106682220A (zh) * 2017-01-04 2017-05-17 华南理工大学 一种基于深度学习的在线中医文本命名实体识别方法
CN107301246A (zh) * 2017-07-14 2017-10-27 河北工业大学 基于超深卷积神经网络结构模型的中文文本分类方法
CN107608964A (zh) * 2017-09-13 2018-01-19 上海六界信息技术有限公司 基于弹幕的直播内容的筛选方法、装置、设备及存储介质
CN107679199A (zh) * 2017-10-11 2018-02-09 北京邮电大学 一种基于深度局部特征的对外汉语教材可读性分析方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103440864A (zh) * 2013-07-31 2013-12-11 湖南大学 基于语音的人格特征预测方法
US9070088B1 (en) * 2014-09-16 2015-06-30 Trooly Inc. Determining trustworthiness and compatibility of a person
CN105630970A (zh) * 2015-12-24 2016-06-01 哈尔滨工业大学 社会媒体数据处理系统及方法
CN106980663A (zh) * 2017-03-21 2017-07-25 上海星红桉数据科技有限公司 基于海量跨屏行为数据的用户画像方法
CN107609009B (zh) * 2017-07-26 2020-02-18 北京大学深圳研究院 文本情感分析方法、装置、存储介质和计算机设备

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106682220A (zh) * 2017-01-04 2017-05-17 华南理工大学 一种基于深度学习的在线中医文本命名实体识别方法
CN107301246A (zh) * 2017-07-14 2017-10-27 河北工业大学 基于超深卷积神经网络结构模型的中文文本分类方法
CN107608964A (zh) * 2017-09-13 2018-01-19 上海六界信息技术有限公司 基于弹幕的直播内容的筛选方法、装置、设备及存储介质
CN107679199A (zh) * 2017-10-11 2018-02-09 北京邮电大学 一种基于深度局部特征的对外汉语教材可读性分析方法

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111581923A (zh) * 2020-04-29 2020-08-25 北京字节跳动网络技术有限公司 文案生成方法、装置、设备和计算机可读存储介质
CN113837445A (zh) * 2021-08-27 2021-12-24 合肥工业大学 基于注意力机制的人格预测方法及系统

Also Published As

Publication number Publication date
CN108829668A (zh) 2018-11-16
CN108829668B (zh) 2021-11-16

Similar Documents

Publication Publication Date Title
CN112732911B (zh) 基于语义识别的话术推荐方法、装置、设备及存储介质
US10049103B2 (en) Author personality trait recognition from short texts with a deep compositional learning approach
US20220337538A1 (en) Customized message suggestion with user embedding vectors
CN111368996A (zh) 可传递自然语言表示的重新训练投影网络
US9672467B2 (en) Systems and methods for creating and implementing an artificially intelligent agent or system
WO2019227629A1 (zh) 文本信息的生成方法、装置、计算机设备及存储介质
JP2020520492A (ja) 文書要約自動抽出方法、装置、コンピュータ機器及び記憶媒体
US11468241B2 (en) Techniques to add smart device information to machine learning for increased context
US10902189B2 (en) System and method to minimally reduce characters in character limiting scenarios
US11727211B2 (en) Systems and methods for colearning custom syntactic expression types for suggesting next best correspondence in a communication environment
US11720757B2 (en) Example based entity extraction, slot filling and value recommendation
US11847423B2 (en) Dynamic intent classification based on environment variables
US11270082B2 (en) Hybrid natural language understanding
CN112995414B (zh) 基于语音通话的行为质检方法、装置、设备及存储介质
CN115017288A (zh) 模型训练方法、模型训练装置、设备及存储介质
CN111680161A (zh) 一种文本处理方法、设备以及计算机可读存储介质
CN114625855A (zh) 用于生成对话信息的方法及装置、设备和介质
CN111639162A (zh) 信息交互方法和装置、电子设备和存储介质
CN116821299A (zh) 智能问答方法、智能问答装置、设备及存储介质
US20220093091A1 (en) Modification of voice commands based on sensitivity
US20230351121A1 (en) Method and system for generating conversation flows
CN109145115B (zh) 产品舆情发现方法、装置、计算机设备和存储介质
CN114239602A (zh) 会话方法、装置和计算机程序产品
JP7099254B2 (ja) 学習方法、学習プログラム及び学習装置
JP7352249B1 (ja) 情報処理装置、情報処理システム及び情報処理方法

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: 18920464

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 26.03.2021)

122 Ep: pct application non-entry in european phase

Ref document number: 18920464

Country of ref document: EP

Kind code of ref document: A1