WO2018000270A1 - 一种基于用户画像的个性化回答生成方法及系统 - Google Patents

一种基于用户画像的个性化回答生成方法及系统 Download PDF

Info

Publication number
WO2018000270A1
WO2018000270A1 PCT/CN2016/087755 CN2016087755W WO2018000270A1 WO 2018000270 A1 WO2018000270 A1 WO 2018000270A1 CN 2016087755 W CN2016087755 W CN 2016087755W WO 2018000270 A1 WO2018000270 A1 WO 2018000270A1
Authority
WO
WIPO (PCT)
Prior art keywords
answer
user
module
candidate
vector
Prior art date
Application number
PCT/CN2016/087755
Other languages
English (en)
French (fr)
Inventor
杨新宇
王昊奋
邱楠
Original Assignee
深圳狗尾草智能科技有限公司
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 深圳狗尾草智能科技有限公司 filed Critical 深圳狗尾草智能科技有限公司
Priority to PCT/CN2016/087755 priority Critical patent/WO2018000270A1/zh
Priority to CN201680001748.0A priority patent/CN106663131A/zh
Publication of WO2018000270A1 publication Critical patent/WO2018000270A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR 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
    • G06F16/33—Querying
    • G06F16/332—Query formulation
    • G06F16/3329—Natural language query formulation or dialogue systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/048—Interaction techniques based on graphical user interfaces [GUI]
    • G06F3/0481—Interaction techniques based on graphical user interfaces [GUI] based on specific properties of the displayed interaction object or a metaphor-based environment, e.g. interaction with desktop elements like windows or icons, or assisted by a cursor's changing behaviour or appearance

Definitions

  • the present invention relates to the field of data processing technologies, and in particular, to a personalized answer generation method and system based on a user portrait.
  • the user portrait also known as the user role (Persona) is an effective tool for sketching the target user, contacting the user's appeal and design direction.
  • the user portrait can be used to locate and plan the product; in the specific implementation, the user portrait can be used as a collection of tags for characterizing the user, such as basic attributes such as age, gender, education, or user interest. Features, etc.; during product promotion, potential customer groups can be mined based on user images for targeted product recommendations.
  • user portraits are gradually being applied to more fields.
  • question-and-answer systems based on artificial intelligence and natural language processing are emerging, allowing users to ask questions in natural language (complete and colloquial questions) and serve users. Return a simple, accurate answer.
  • the existing questions and answers are generated by the question and answer, and the user experience is not rich enough for the user to be personalized.
  • the present invention provides a personalized answer generation method and system based on a user portrait, which combines a user portrait vector with a traditional answer vector to obtain a personalized answer generation and realize the question and answer personality.
  • a personalized answer generation method and system based on a user portrait, which combines a user portrait vector with a traditional answer vector to obtain a personalized answer generation and realize the question and answer personality.
  • the present invention provides a personalized answer generation method based on a user portrait, comprising: Step 1: A user inputs a multimodal input, and performs multimodal input conversion on the multimodal input, and Convert to text question; Step 2: Answer the generated answer based on the converted text question, and score each candidate answer generated; Step 3: Generate the user image The vector and the vector of each candidate answer are similarly calculated; Step 4: weighting and rearranging the similarity calculated in step 3 and the score obtained in step 2, and selecting the candidate with the highest score As the answer to the final output.
  • said step 2-4 is performed on a server.
  • the present invention also provides a personalized answer generation method based on a user portrait, comprising: step 1: a user inputting a text question; step 2: answering according to the text question to generate a candidate answer, and generating each of the The candidate answer is scored; step 3: the vector generated by the user image and the vector of each candidate answer are similarly calculated; step 4: weighting the similarity calculated in step 3 with the score obtained in step 2 Add, rearrange, and use the highest-scoring candidate answer as the answer to the final output.
  • said step 2-4 is performed on a server.
  • the present invention also provides a personalized answer generation system based on a user portrait, comprising: a multi-modal input conversion module for converting a multi-modal input input by a user into a multi-modal input into a text problem.
  • An answer generation module configured to generate an answer according to the converted text question, and score each generated candidate answer;
  • a user portrait similarity calculation module configured to generate a vector of the user image and each candidate answer The vector performs similarity calculation;
  • an answer output module is configured to weight-add and rearrange the calculated similarity with the score obtained by the answer generation module, and output the candidate answer with the highest score as the final answer.
  • the answer generation module, the user portrait similarity calculation module, and the answer output module are on a server.
  • the present invention further provides a personalized answer generation system based on a user portrait, comprising: a question input module, an answer generation module, a user portrait similarity calculation module, and an answer output module.
  • the problem input module is configured to receive a question input by the user and send the question to the question and answer generation module;
  • the answer generation module is configured to generate an answer according to the input text question, and score each candidate answer generated;
  • the similarity of the user image a calculation module for performing a similarity calculation on a vector generated by the user image and a vector of each candidate answer;
  • an answer output module for The similarity of the calculation is used as another score, and the scores obtained by the traditional method are weighted and rearranged, and the candidate answers with the highest score are output as the final answer.
  • the answer generation module, the user portrait similarity calculation module, and the answer output module are on a server.
  • the answer generation generated by the user's portrait can satisfy the individualized characteristics
  • FIG. 1 is a flowchart of a personalized answer generation method based on a user image according to an embodiment of the present invention
  • FIG. 2 is a schematic structural diagram of a personalized answer generation system based on a user portrait according to an embodiment of the present invention
  • FIG. 3 is a schematic structural diagram of a personalized answer generation system based on a user portrait according to another embodiment of the present invention.
  • FIG. 1 is a flowchart of a method for generating a personalized answer based on a user portrait according to an embodiment of the present invention, including the following steps:
  • Step 1 The user inputs the multi-modal input and performs multi-modal input conversion on the input multi-modal input. Turn it into a text problem. For example, a user enters a voice and performs a voice conversion on the input voice to turn it into a text question.
  • multimodal input includes, but is not limited to, video, face, expression, scene, voiceprint, fingerprint, iris pupil, light perception, and the like.
  • Step 2 Answer the generated answer based on the converted text question and score each candidate answer generated.
  • the answer generation may be performed using a traditional answer generation method
  • Step 3 Perform a similarity calculation on the vector generated by the user image and the vector of each candidate answer
  • Step 4 The similarity calculated in the above step 3 is taken as another score, and the scores obtained by the traditional method are weighted and added, rearranged, and the candidate answer with the highest score is used as the final output answer.
  • the user can directly input the text question, and the steps 1-4 can be performed at the server.
  • the steps of inputting voice and voice conversion are not necessary.
  • FIG. 2 is a schematic structural diagram of a personalized answer generation system based on a user portrait according to an embodiment of the present invention, including a multi-modal input conversion module (eg, a voice input conversion module), an answer generation module, and a user portrait similarity calculation. Module and answer output module.
  • the voice input conversion module is configured to convert the voice input by the user into a text question, and convert the voice into a text question;
  • the answer generation module is configured to generate an answer according to the converted text question, and score each candidate answer generated.
  • a user portrait similarity calculation module for performing a similarity calculation on a vector generated by the user image and a vector of each candidate answer; and an answer output module for using the calculated similarity as another score, which is obtained in a conventional manner
  • the scores are weighted and reordered, and the candidate answers with the highest score are output as the final answer.
  • FIG. 3 is a schematic structural diagram of a personalized answer generation system based on a user portrait according to another embodiment of the present invention, including a question input module, an answer generation module, a user portrait similarity calculation module, and an answer output module.
  • the problem input module is configured to receive a question input by the user and send the question to the question and answer generation module;
  • the answer generation module is configured to generate an answer according to the input text question, and score each candidate answer generated; the similarity of the user image a calculation module for performing a similarity calculation on a vector generated by the user image and a vector of each candidate answer; and an answer output module,
  • the calculated similarity is taken as another score, and the scores obtained by the traditional method are weighted and rearranged, and the candidate answers with the highest score are output as the final answer.
  • the answer generation module, the user portrait similarity calculation module, and the answer output module are on a server.
  • the personalized answer generation method and system based on the user portrait provided by the present invention realizes the answer generation in the question and answer by means of the user portrait, and can satisfy the function of the personalized answer.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Human Computer Interaction (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Mathematical Physics (AREA)
  • Computational Linguistics (AREA)
  • Artificial Intelligence (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Electrically Operated Instructional Devices (AREA)

Abstract

一种基于用户画像的个性化回答生成方法及系统，该方法包括：步骤1：用户输入多模态输入，并对所述多模态输入进行多模态输入转化，将其转化为文本问题；步骤2：根据所述转化的文本问题进行回答生成候选答案，并对生成的每一个候选答案进行打分；步骤3：将用户画像生成的向量和每一个候选答案的向量进行相似度计算；步骤4：将所述步骤3计算的相似度与所述步骤2得出的分数进行加权相加、重排，并将得分最高的候选回答作为最终输出的答案。通过用户画像的方式来实现问答中的答案生成，能满足个性化回答的功能。

Description

一种基于用户画像的个性化回答生成方法及系统 技术领域
本发明涉及数据处理技术领域，并且特别涉及一种基于用户画像的个性化回答生成方法及系统。
背景技术
用户画像，又称为用户角色(Persona)，是一种勾画目标用户、联系用户诉求与设计方向的有效工具。例如在产品开发时，可用于对产品进行定位与规划；在具体实现时，可以将用户画像作为刻画用户特征的标签(tag)集合，例如：年龄、性别、学历等基础属性，或者用户的兴趣特征等；在产品推广时，可根据用户画像挖掘潜在客户群体，进行有针对性的产品推荐。随着信息技术的不断发展，用户画像也逐渐应用于更多领域中。
随着人们对快速、准确地获取信息的需求不断增加，基于人工智能和自然语言处理领域的问答系统逐渐兴起，其能让用户用自然语言提问(完整而口语化的问句)，并为用户返回一个简洁、准确的答案。但现有的问答都是通过问句来进行回答的判断生成，对于用户没有个性化，用户体验不够丰富。
发明内容
针对现有技术的不足，本发明提供一种基于用户画像的个性化回答生成方法及系统，采用将用户画像向量与传统回答向量相结合的方式，得到具有个性化的回答生成，实现问答的个性化体验。
为解决上述技术问题，本发明提供一种基于用户画像的个性化回答生成方法，包括：步骤1：用户输入多模态输入，并对所述多模态输入进行多模态输入转化，将其转化为文本问题；步骤2：根据所述转化的文本问题进行回答生成候选答案，并对生成的每一个候选答案进行打分；步骤3：将用户画像生成 的向量和每一个候选答案的向量进行相似度计算；步骤4：将所述步骤3计算的相似度与所述步骤2得出的分数进行加权相加、重排，并将得分最高的候选回答作为最终输出的答案。
优选地，所述步骤2-4是在服务器上执行。
本发明还提供一种基于用户画像的个性化回答生成方法，其特征在于，包括：步骤1：用户输入文本问题；步骤2：根据所述文本问题进行回答生成候选答案，并对生成的每一个候选答案进行打分；步骤3：将用户画像生成的向量和每一个候选答案的向量进行相似度计算；步骤4：将所述步骤3计算的相似度与所述步骤2得出的分数进行加权相加、重排，并将得分最高的候选回答作为最终输出的答案。
优选地，所述步骤2-4是在服务器上执行。
为解决上述技术问题，本发明还提供一种基于用户画像的个性化回答生成系统，包括：多模态输入转化模块，用于将用户输入的多模态输入进行多模态输入转化为文本问题；回答生成模块，用于根据所述转化的文本问题进行回答生成，并对生成的每一个候选答案进行打分；用户画像相似度计算模块，用于将用户画像生成的向量和每一个候选答案的向量进行相似度计算；以及答案输出模块，用于将所述计算的相似度与所述回答生成模块得出的分数进行加权相加、重排，并将得分最高的候选回答作为最终答案输出。
优选地，所述回答生成模块、用户画像相似度计算模块、和答案输出模块是在服务器上。
为解决上述技术问题，本发明还提供一种基于用户画像的个性化回答生成系统，包括：问题输入模块、回答生成模块、用户画像相似度计算模块以及答案输出模块。其中，问题输入模块，用于接收用户输入的问题并发送至问答生成模块；回答生成模块，用于根据输入的文本问题进行回答生成，并对生成的每一个候选答案进行打分；用户画像相似度计算模块，用于将用户画像生成的向量和每一个候选答案的向量进行相似度计算；以及答案输出模块，用于将计 算的相似度作为另一个分数，与传统方式得出的分数进行加权相加、重排，并将得分最高的候选回答作为最终答案输出。
优选地，所述回答生成模块、用户画像相似度计算模块、和答案输出模块是在服务器上。
总体而言，相较于现有技术，本发明的技术方案具有以下有益效果：
1、通过用户画像进行回答生成能满足个性化的特点；
2、将用户画像向量与候选答案向量进行相似度计算，并将计算的相似度与传统方式得到的答案分数进行加权计算，得到的回答更准确。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本发明一实施例提供的基于用户画像的个性化回答生成方法的流程图；
图2是本发明一实施例提供的基于用户画像的个性化回答生成系统的结构示意图；
图3是本发明另一实施例提供的基于用户画像的个性化回答生成系统的结构示意图。
具体实施方式
为了使本发明的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本发明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仅用以解释本发明，并不用于限定本发明。
图1所示为本发明一实施例提供的基于用户画像的个性化回答生成方法的流程图，包括以下步骤：
步骤1：用户输入多模态输入，并对输入的多模态输入进行多模态输入转 化，将其转化为文本问题。例如，用户输入语音，并对输入的语音进行语音转化，将其转化为文本问题。请注意，本文所指的“多模态输入”包括但不限于，视频、人脸、表情、场景、声纹、指纹、虹膜瞳孔、光感、等信息。
步骤2：根据转化的文本问题进行回答生成候选答案，并对生成的每一个候选答案进行打分。在本发明实施例中，可使用传统的回答生成方式进行回答生成；
步骤3：将用户画像生成的向量和每一个候选答案的向量进行相似度计算；
步骤4：将上述步骤3计算的相似度作为另一个分数，与传统方式得出的分数进行加权相加、重排，并将得分最高的候选回答作为最终输出的答案。
其中，所述步骤1中，用户也可以直接输入文本问题，并且步骤1-4可在服务器执行。也就是说，输入语音以及语音转化的步骤并非是必要的。
图2所示为本发明一实施例提供的基于用户画像的个性化回答生成系统的结构示意图，包括多模态输入转化模块(例如，语音输入转化模块)、回答生成模块、用户画像相似度计算模块以及答案输出模块。其中，语音输入转化模块，用于将用户输入的语音进行语音转化，将其转化为文本问题；回答生成模块，用于根据转化的文本问题进行回答生成，并对生成的每一个候选答案进行打分；用户画像相似度计算模块，用于将用户画像生成的向量和每一个候选答案的向量进行相似度计算；以及答案输出模块，用于将计算的相似度作为另一个分数，与传统方式得出的分数进行加权相加、重排，并将得分最高的候选回答作为最终答案输出。
图3所示为本发明另一实施例提供的基于用户画像的个性化回答生成系统的结构示意图，包括问题输入模块、回答生成模块、用户画像相似度计算模块以及答案输出模块。其中，问题输入模块，用于接收用户输入的问题并发送至问答生成模块；回答生成模块，用于根据输入的文本问题进行回答生成，并对生成的每一个候选答案进行打分；用户画像相似度计算模块，用于将用户画像生成的向量和每一个候选答案的向量进行相似度计算；以及答案输出模块，用 于将计算的相似度作为另一个分数，与传统方式得出的分数进行加权相加、重排，并将得分最高的候选回答作为最终答案输出。
在一个实施例中，所述回答生成模块、用户画像相似度计算模块、和答案输出模块是在服务器上。
本发明提供的基于用户画像的个性化回答生成方法及系统，通过用户画像的方式来实现问答中的答案生成，能满足个性化回答的功能。
以上所述仅为本发明的较佳实施例而已，并不用以限制本发明，凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等，均应包含在本发明的保护范围之内。

Claims (8)

  1. 一种基于用户画像的个性化回答生成方法，其特征在于，包括：
    步骤1：用户输入多模态输入，并对所述多模态输入进行多模态输入转化，将其转化为文本问题；
    步骤2：根据所述转化的文本问题进行回答生成候选答案，并对生成的每一个候选答案进行打分；
    步骤3：将用户画像生成的向量和每一个候选答案的向量进行相似度计算；
    步骤4：将所述步骤3计算的相似度与所述步骤2得出的分数进行加权相加、重排，并将得分最高的候选回答作为最终输出的答案。
  2. 如权利要求1所述的基于用户画像的个性化回答生成方法，其特征在于，所述步骤2-4是在服务器上执行。
  3. 一种基于用户画像的个性化回答生成方法，其特征在于，包括：
    步骤1：用户输入文本问题；
    步骤2：根据所述文本问题进行回答生成候选答案，并对生成的每一个候选答案进行打分；
    步骤3：将用户画像生成的向量和每一个候选答案的向量进行相似度计算；
    步骤4：将所述步骤3计算的相似度与所述步骤2得出的分数进行加权相加、重排，并将得分最高的候选回答作为最终输出的答案。
  4. 如权利要求3所述的基于用户画像的个性化回答生成方法，其特征在于，所述步骤2-4是在服务器上执行。
  5. 一种基于用户画像的个性化回答生成系统，其特征在于，包括：
    多模态输入转化模块，用于将用户输入的多模态输入进行多模态输入转化为文本问题；
    回答生成模块，用于根据所述转化的文本问题进行回答生成，并对生成的每一个候选答案进行打分；
    用户画像相似度计算模块，用于将用户画像生成的向量和每一个候选答案 的向量进行相似度计算；以及
    答案输出模块，用于将所述计算的相似度与所述回答生成模块得出的分数进行加权相加、重排，并将得分最高的候选回答作为最终答案输出。
  6. 如权利要求5所述的基于用户画像的个性化回答生成系统，其特征在于，所述回答生成模块、用户画像相似度计算模块、和答案输出模块是在服务器上。
  7. 一种基于用户画像的个性化回答生成系统，其特征在于，包括：
    问题输入模块，用于接收用户输入的文本问题；
    回答生成模块，用于根据所述文本问题进行回答生成，并对生成的每一个候选答案进行打分；
    用户画像相似度计算模块，用于将用户画像生成的向量和每一个候选答案的向量进行相似度计算；以及
    答案输出模块，用于将所述计算的相似度与所述回答生成模块得出的分数进行加权相加、重排，并将得分最高的候选回答作为最终答案输出。
  8. 如权利要求7所述的基于用户画像的个性化回答生成系统，其特征在于，所述回答生成模块、用户画像相似度计算模块、和答案输出模块是在服务器上。
PCT/CN2016/087755 2016-06-29 2016-06-29 一种基于用户画像的个性化回答生成方法及系统 WO2018000270A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2016/087755 WO2018000270A1 (zh) 2016-06-29 2016-06-29 一种基于用户画像的个性化回答生成方法及系统
CN201680001748.0A CN106663131A (zh) 2016-06-29 2016-06-29 一种基于用户画像的个性化回答生成方法及系统

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2016/087755 WO2018000270A1 (zh) 2016-06-29 2016-06-29 一种基于用户画像的个性化回答生成方法及系统

Publications (1)

Publication Number Publication Date
WO2018000270A1 true WO2018000270A1 (zh) 2018-01-04

Family

ID=58838407

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/087755 WO2018000270A1 (zh) 2016-06-29 2016-06-29 一种基于用户画像的个性化回答生成方法及系统

Country Status (2)

Country Link
CN (1) CN106663131A (zh)
WO (1) WO2018000270A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110109961A (zh) * 2019-05-10 2019-08-09 宜信博诚保险销售服务(北京)股份有限公司 用户画像获取方法及装置、用户需求获取方法及装置

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108875055B (zh) * 2018-06-28 2021-12-24 联想(北京)有限公司 一种答案提供方法及设备
CN109376361A (zh) * 2018-11-16 2019-02-22 北京九狐时代智能科技有限公司 一种意图识别方法及装置
CN110457440B (zh) * 2019-08-09 2022-08-16 宝宝树(北京)信息技术有限公司 一种反馈答案的方法、装置、设备和介质
CN110674276A (zh) * 2019-09-23 2020-01-10 深圳前海微众银行股份有限公司 机器人自学习方法、机器人终端、装置及可读存储介质
CN111741104B (zh) * 2020-06-18 2021-10-08 腾讯科技（深圳）有限公司 应答消息的确定方法、配置方法、装置、设备及存储介质
CN111858716B (zh) * 2020-07-24 2023-10-13 厦门至恒融兴信息技术有限公司 海量金融支付数据的自动流程化处理技术
CN115190336B (zh) * 2022-06-27 2023-02-28 青软创新科技集团股份有限公司 一种面向众包式网络教学的直播连接方法及系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103729424A (zh) * 2013-12-20 2014-04-16 百度在线网络技术（北京）有限公司 问答社区内回答评价方法和系统
CN105095357A (zh) * 2015-06-24 2015-11-25 百度在线网络技术（北京）有限公司 一种用于咨询数据处理的方法和装置
CN105183803A (zh) * 2015-08-25 2015-12-23 天津大学 一种社交网络平台中的个性化搜索方法及其搜索装置
CN105243143A (zh) * 2015-10-14 2016-01-13 湖南大学 基于即时语音内容检测的推荐方法及系统

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103761286B (zh) * 2014-01-14 2017-03-08 河南科技大学 一种基于用户兴趣的服务资源检索方法
CN105068661B (zh) * 2015-09-07 2018-09-07 百度在线网络技术（北京）有限公司 基于人工智能的人机交互方法和系统
CN105487663B (zh) * 2015-11-30 2018-09-11 北京光年无限科技有限公司 一种面向智能机器人的意图识别方法和系统
CN105574133A (zh) * 2015-12-15 2016-05-11 苏州贝多环保技术有限公司 一种多模态的智能问答系统及方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103729424A (zh) * 2013-12-20 2014-04-16 百度在线网络技术（北京）有限公司 问答社区内回答评价方法和系统
CN105095357A (zh) * 2015-06-24 2015-11-25 百度在线网络技术（北京）有限公司 一种用于咨询数据处理的方法和装置
CN105183803A (zh) * 2015-08-25 2015-12-23 天津大学 一种社交网络平台中的个性化搜索方法及其搜索装置
CN105243143A (zh) * 2015-10-14 2016-01-13 湖南大学 基于即时语音内容检测的推荐方法及系统

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110109961A (zh) * 2019-05-10 2019-08-09 宜信博诚保险销售服务(北京)股份有限公司 用户画像获取方法及装置、用户需求获取方法及装置

Also Published As

Publication number Publication date
CN106663131A (zh) 2017-05-10

Similar Documents

Publication Publication Date Title
WO2018000270A1 (zh) 一种基于用户画像的个性化回答生成方法及系统
US10642887B2 (en) Multi-modal image ranking using neural networks
WO2018000271A1 (zh) 一种基于用户画像的意图场景识别方法及系统
WO2021134277A1 (zh) 情感识别方法、智能装置和计算机可读存储介质
Gharavian et al. Audio-visual emotion recognition using FCBF feature selection method and particle swarm optimization for fuzzy ARTMAP neural networks
CN113792177B (zh) 基于知识引导深度注意力网络的场景文字视觉问答方法
WO2021174922A1 (zh) 语句情感分类方法及相关设备
US20210232751A1 (en) Sentence generation method and apparatus, device, and storage medium
US11080073B2 (en) Computerized task guidance across devices and applications
KR20220064940A (ko) 음성 생성 방법, 장치, 전자기기 및 저장매체
KR102121654B1 (ko) 딥러닝 기반 제스처 자동 인식 방법 및 시스템
Camgöz et al. Sign language recognition for assisting the deaf in hospitals
CN113096242A (zh) 虚拟主播生成方法、装置、电子设备及存储介质
CN112685550B (zh) 智能问答方法、装置、服务器及计算机可读存储介质
CN113761220A (zh) 信息获取方法、装置、设备及存储介质
WO2023134069A1 (zh) 实体关系的识别方法、设备及可读存储介质
Wang et al. (2+ 1) D-SLR: an efficient network for video sign language recognition
Minato et al. Design of a competition specifically for spoken dialogue with a humanoid robot
CN111444321B (zh) 问答方法、装置、电子设备和存储介质
CN117315334A (zh) 图像分类方法、模型的训练方法、装置、设备及介质
US9524656B2 (en) Sign language image input method and device
WO2023154351A2 (en) Apparatus and method for automated video record generation
CN115017900B (zh) 一种基于多模态多去偏见的对话情感识别方法
Kaimakis et al. Domain-specific reduction of language model databases: Overcoming chatbot implementation obstacles
CN113763925B (zh) 语音识别方法、装置、计算机设备及存储介质

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: 16906671

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: 16906671

Country of ref document: EP

Kind code of ref document: A1