WO2018028319A1 - 一种联系人排序方法和装置 - Google Patents

一种联系人排序方法和装置 Download PDF

Info

Publication number
WO2018028319A1
WO2018028319A1 PCT/CN2017/090369 CN2017090369W WO2018028319A1 WO 2018028319 A1 WO2018028319 A1 WO 2018028319A1 CN 2017090369 W CN2017090369 W CN 2017090369W WO 2018028319 A1 WO2018028319 A1 WO 2018028319A1
Authority
WO
WIPO (PCT)
Prior art keywords
word segmentation
text
contact
content
text content
Prior art date
Application number
PCT/CN2017/090369
Other languages
English (en)
French (fr)
Inventor
左焘
Original Assignee
中兴通讯股份有限公司
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 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2018028319A1 publication Critical patent/WO2018028319A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M1/00—Substation equipment, e.g. for use by subscribers
    • H04M1/26—Devices for calling a subscriber
    • H04M1/27—Devices whereby a plurality of signals may be stored simultaneously
    • H04M1/274—Devices whereby a plurality of signals may be stored simultaneously with provision for storing more than one subscriber number at a time, e.g. using toothed disc
    • H04M1/2745—Devices whereby a plurality of signals may be stored simultaneously with provision for storing more than one subscriber number at a time, e.g. using toothed disc using static electronic memories, e.g. chips
    • H04M1/27453—Directories allowing storage of additional subscriber data, e.g. metadata
    • H04M1/2746—Sorting, e.g. according to history or frequency of use
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M1/00—Substation equipment, e.g. for use by subscribers
    • H04M1/26—Devices for calling a subscriber
    • H04M1/27—Devices whereby a plurality of signals may be stored simultaneously
    • H04M1/274—Devices whereby a plurality of signals may be stored simultaneously with provision for storing more than one subscriber number at a time, e.g. using toothed disc
    • H04M1/2745—Devices whereby a plurality of signals may be stored simultaneously with provision for storing more than one subscriber number at a time, e.g. using toothed disc using static electronic memories, e.g. chips
    • H04M1/27453—Directories allowing storage of additional subscriber data, e.g. metadata
    • H04M1/27457—Management thereof, e.g. manual editing of data

Definitions

  • This document relates to, but is not limited to, the field of communication technologies, and in particular, to a method and device for sorting contacts.
  • the embodiment of the invention provides a method and device for sorting contacts, which can improve the efficiency of contact searching.
  • An embodiment of the present invention provides a method for sorting contacts, including:
  • the contacts in the address book are sorted according to the emotional score of each contact.
  • An embodiment of the present invention further provides a contact sorting apparatus, including:
  • An acquisition module configured to obtain an emotional score for each contact in the address book
  • a sorting module is arranged to sort the contacts in the address book according to the emotional score of each contact.
  • Embodiments of the present invention also provide a computer storage medium, where the computer storage medium is stored There is one or more programs executable by the computer, the one or more programs being executed by the computer to cause the computer to perform a contact sorting method as provided above.
  • the emotional scores of each contact in the address book are obtained; the contacts in the address book are sorted according to the emotional score of each contact, so that the contact that needs to be found can be quickly found.
  • FIG. 1 is a flowchart of a method for sorting contacts according to an embodiment of the present invention
  • FIG. 3 is a schematic diagram of a contact sorting apparatus according to an embodiment of the present invention.
  • FIG. 5 is another contact sorting apparatus according to an embodiment of the present invention.
  • FIG. 6 is another contact sorting apparatus according to an embodiment of the present invention.
  • FIG. 7 is another device for sorting contacts according to an embodiment of the present invention.
  • FIG. 1 is a flowchart of a method for sorting contacts according to an embodiment of the present invention, including the following steps:
  • Step S101 Acquire an emotional score of each contact in the address book.
  • Each contact in the address book may be each contact in the address book corresponding to the communication account, and the address book may be used to record the contact information, and the information may include the contact address of the contact, and the contact information.
  • Name, contact's communication account, email address, etc., the above communication account may include an instant messaging account
  • each of the above contacts may be understood as each contact added in the above communication account
  • the contact's communication account may be added.
  • the emotional score of each of the above contacts may be the emotional score of each of the above contacts obtained through the communication account and the communication record of each contact.
  • the communication record of the communication account and each contact may be a communication record generated when the communication account is contacted with each contact, and the communication record may be text, picture, voice, and other forms, and the communication record may include more
  • the language may include a plurality of chat modes, such as a dialog box chat or a message, etc., and the content and manner of the communication are not limited in this embodiment.
  • the emotional score may be based on the communication record, determining the emotional polarity of the communication record, and calculating the emotional score of the communication record, thereby calculating the emotional score of the contact.
  • Obtaining the emotional score of each contact in the address book may be obtaining the emotional score of each contact in the address book separately.
  • Step S102 Sort the contacts in the address book according to the emotional score of each contact.
  • the step may be to sort the sentiment scores of each contact in the address book obtained in step S101, which may be sorted according to the scores from high to low or low to high, or may be sorted according to other rules.
  • the contacts corresponding to the emotional scores are simultaneously sorted. For example, there are contacts A, B, C, D, and E in the address book. If the emotional scores of the contacts are ranked C, B, A, E, and D from high to low, then the contact is based on the emotional score of the contact.
  • the result of sorting is C, B, A, E, D.
  • the embodiments of the present invention can be applied to terminals, such as a mobile phone, a tablet personal computer, a laptop computer, a personal digital assistant (PDA), and a wearable device. Wait.
  • terminals such as a mobile phone, a tablet personal computer, a laptop computer, a personal digital assistant (PDA), and a wearable device. Wait.
  • PDA personal digital assistant
  • the emotional scores of each contact in the address book are obtained; the contacts in the address book are sorted according to the emotional score of each contact, so that the contact that needs to be found can be quickly found.
  • FIG. 2 is a flowchart of another method for sorting contacts according to an embodiment of the present invention.
  • the address book includes a target contact, and the method includes the following steps:
  • Step S201 Acquire a communication record of the target contact.
  • the target contact may be any one of the contacts in the address book. It should be noted that the target contact may be any one of the contact groups that include multiple contacts in the address book, and the target contact is included in the multiple contacts, and the target contact is sent to the target contact. Score calculation, of course, the contacts in the above address book, in addition to the above target contacts, other contacts also perform the above steps of emotional score calculation. For the sake of clarity, any one of the above contacts, ie, the above-mentioned target contact, is selected here to explain the process of calculating the emotional score.
  • the foregoing communication record may be a chat record of the communication account and the target contact, and the content of the chat record may include a text, a picture, an audio, and the like, which is not limited in this embodiment.
  • the obtaining the communication record of the target contact may be obtaining a communication record sent to the target contact by using the communication account.
  • Step S202 Perform word segmentation on the communication record.
  • the step may be to segment the content of the communication record to decompose the content of the communication record into a plurality of individual words.
  • the above content can be converted into a preset form. For example, if a language is expressed by means of pictures, animations, etc., the above-mentioned pictures, animations, and the like may be converted into text content having the same or similar meaning; for the above-mentioned communication record text content containing slang, the content expressed in slang may be expressed. Convert to written language and more.
  • Step S203 Calculate an emotional score of the text content after the word segmentation, and use the emotion score as the emotional score of the target contact.
  • the emotional score of the text content after the word segmentation may be determined according to the emotional polarity of the text content after the communication record segmentation in step S202, and the emotional score of the text content after the segmentation is calculated.
  • the emotional score of the text content after the above-mentioned word segmentation may be calculated by using a trained naive Bayesian algorithm, and the calculated emotional score may be the emotional score of the target contact.
  • Step S204 Sort the contacts in the address book according to the emotional score of each contact.
  • the segmentation of the communication record includes:
  • the word content is segmented.
  • the text content that converts the communication content into the preset condition may be a text content that converts the communication content into a format, a language, an expression, and the like to satisfy a preset condition, so that the communication content is more easily recognized. And the identification is more accurate.
  • the format of the text content in the text content that is traditionally input is converted into a simplified format, and the format of the text content output in the full-width mode in the input method is converted into the format output in the half-width mode; the slang in the text content can be converted.
  • the text content contains “cups”, which can be converted into “tragedy”; the expressions or texts input in the communication content by means of pictures can be converted into the text content corresponding to the expression or the facial expression.
  • the content of the above-mentioned conversion may also be other situations, which are merely examples, which are not limited in this embodiment.
  • the text content of the converted communication content is segmented.
  • the communication content in the communication record is converted into the text content that satisfies the preset condition, and the text content is segmented, so that the communication content can be better recognized and the recognition is more accurate.
  • the segmentation of the text content includes:
  • the recognized character may be a recognized universal keyword, for example, identifying a person name, a date, a time, a phone number, and the like in the content of the communication record.
  • the above identification may be to identify the text by using a regular expression, and the The above-identified characters use the BIES word segmentation method or other methods to output the word segmentation information of the recognized characters.
  • BIES word segmentation label "Beijing” can be identified, and the word segmentation labels of "North” and "Beijing" are "B” and "E", respectively. This is only an example, and the embodiment does not limit this.
  • the texts that cannot be recognized in the above text content can be trained by using a toolkit such as word2vec to obtain a word vector, or a random vector can be used to obtain a word vector.
  • the above word vector can be operated using a trained network model including cnn convolution.
  • the neural network model, the dnn deep learning network model, and the like, output word segmentation label information of the text that cannot be recognized in the text content.
  • the word segmentation labels of the texts in the above text content are respectively assigned scores, the label score vectors of each word are merged into a score matrix, the column represents words, the rows represent labels, and the Viterbi algorithm is used to perform operations, and finally the text content of the communication record is finally determined. Segmentation information.
  • the text content in the above communication record is divided into characters that can be recognized and cannot be recognized, and different methods are used to obtain the word segmentation label of the text, which can optimize resources, reduce the amount of calculation, and can recognize and cannot
  • the identified text content is calculated by using the Viterbi algorithm to obtain word segmentation information of the text content, so that the obtained segmentation information is more accurate.
  • the calculating the emotional score of the text content after the word segmentation includes:
  • the emotional score of the textual content of the word segmentation is calculated using a trained naive Bayesian algorithm.
  • the trained naive Bayes algorithm may be a training of the naive Bayes algorithm, and the training may include training the naive Bayes algorithm according to the sentiment dictionary, so that the naive Bayes algorithm can determine the emotional polarity. And calculate emotional scores based on emotional polarity.
  • the text content of the above segmentation information is calculated by the Na ⁇ ve Bayes algorithm, and the score of the emotional polarity is output.
  • An embodiment of the present invention converts a communication record into a text content that satisfies a preset condition on the basis of the corresponding embodiment of FIG. 1, and uses different texts that can be recognized in the text content and text content that cannot be recognized.
  • the method performs word segmentation, finally determines the word segmentation information, calculates the emotion score of the text content according to the word segmentation information, optimizes the resource, and makes the obtained segmentation word information and the emotion score more accurate, so that the user terminal can find the contact person more quickly.
  • FIG. 3 is a contact sorting apparatus 300 according to an embodiment of the present invention.
  • the contact sorting apparatus 300 includes an obtaining module 301 and a sorting module 302.
  • the obtaining module 301 is configured to obtain an emotional score of each contact in the address book
  • the sorting module 302 is configured to connect the contacts in the address book according to the emotional score of each contact The person sorts.
  • the address book includes at least a target contact
  • the obtaining module 301 includes:
  • the obtaining submodule 3011 is configured to obtain a communication record of the target contact
  • a word segmentation sub-module 3012 configured to perform word segmentation on the communication record
  • the calculation sub-module 3013 is configured to calculate an emotional score of the text content after the word segmentation, and use the emotion score as the emotional score of the target contact.
  • the word segmentation sub-module 3012 includes:
  • the converting unit 30121 is configured to convert the communication content in the communication record into text content that meets a preset condition
  • the word segmentation unit 30122 is configured to segment the text content.
  • the word segmentation unit 30122 includes:
  • the obtaining subunit 301221 is configured to acquire word segmentation information of the text that can be recognized in the text content;
  • the first operation subunit 301222 is configured to perform a network model operation on a word vector of the text that cannot be recognized in the text content, and obtain word segmentation information of the text that cannot be recognized in the text content;
  • a second operation subunit 301223 configured to perform a Viterbi operation on the word segmentation information of the recognizable character and the word segmentation information of the unrecognizable character, and determine word segmentation information of the text of the communication record .
  • the calculation submodule 3013 includes:
  • the 30131 computing unit is configured to calculate an emotional score of the textual content of the segmentation using a trained naive Bayesian algorithm.
  • the contact sorting device 300 can be applied to a terminal, such as a mobile phone, a tablet personal computer, a laptop computer, a personal digital assistant (PDA), or a personal digital assistant (PDA). Wearable Device, etc.
  • a terminal such as a mobile phone, a tablet personal computer, a laptop computer, a personal digital assistant (PDA), or a personal digital assistant (PDA).
  • PDA personal digital assistant
  • PDA personal digital assistant
  • Wearable Device etc.
  • the above-mentioned contact sorting device 300 can implement various processes implemented by the terminal in the method embodiment corresponding to FIG. 1 to FIG. 2, and details are not repeatedly described herein.
  • the contact ranking device 300 can sort the contacts in the address book to enable the user terminal to quickly find the contacts.
  • the contacts in the address book are sorted according to the emotional score of each contact.
  • the address book includes at least a target contact
  • the obtaining an emotional score of each contact in the address book includes:
  • the segmentation of the communication record includes:
  • the word content is segmented.
  • the segmentation of the text content includes:
  • the calculating the emotional score of the text content after the word segmentation includes:
  • the emotional score of the textual content of the word segmentation is calculated using a trained naive Bayesian algorithm.
  • computer storage medium includes volatile and nonvolatile, implemented in any method or technology for storing information, such as computer readable instructions, data structures, program modules, or other data. , removable and non-removable media.
  • Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridge, magnetic tape, magnetic disk storage or other magnetic storage device, or may Any other medium used to store the desired information and that can be accessed by the computer.
  • communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal, such as a carrier wave or other transport mechanism, and can include any information delivery media. .
  • the above technical solution can quickly find the contact that needs to be found, simplify the search procedure, and improve the search efficiency.

Abstract

一种联系人排序方法和装置，该方法可包括：获取通讯录中每个联系人的情感分数（S101）；按照每个联系人的情感分数对所述通讯录中的联系人进行排序（S102）。

Description

一种联系人排序方法和装置 技术领域
本文涉及但不限于通信技术领域，特别涉及一种联系人排序方法和装置。
背景技术
随着通讯工具的发展，通讯工具使用用户逐步增多，通讯客户端的通讯录上添加的联系人数量也越来越多。当通讯用户需要查找通讯录中的其中一个联系人时，可以按照拼音或者笔画输入该联系人的称呼进行查找，或输入该联系人的称呼的首字获取相关的排序信息，从而进行查找。但是，这种查找联系人的方式需要执行较多操作，效率较低。
发明内容
以下是对本文详细描述的主题的概述。本概述并非是为了限制权利要求的保护范围。
本发明实施例提供一种联系人排序方法和装置，能够提高联系人查找效率。
本发明实施例提供一种联系人排序方法，包括：
获取通讯录中每个联系人的情感分数；
按照每个联系人的情感分数对所述通讯录中的联系人进行排序。
本发明实施例还提供一种联系人排序装置，包括：
获取模块，设置为获取通讯录中每个联系人的情感分数；
排序模块，设置为按照每个联系人的情感分数对所述通讯录中的联系人进行排序。
本发明实施例还提供一种计算机存储介质，所述计算机存储介质中存储 有计算机可执行的一个或多个程序，所述一个或多个程序被所述计算机执行时使所述计算机执行如上述提供的一种联系人排序方法。
上述技术方案具有如下优点或有益效果：
本发明实施例通过获取通讯录中每个联系人的情感分数；按照每个联系人的情感分数对所述通讯录中的联系人进行排序，这样，能够快速找到需要查找的联系人。
在阅读并理解了附图和详细描述后，可以明白其他方面。
附图概述
图1为本发明实施例提供的一种联系人排序方法的流程图；
图2为本发明实施例提供的另一种联系人排序方法的流程图；
图3为本发明实施例提供的一种联系人排序装置；
图4为本发明实施例提供的另一种联系人排序装置；
图5为本发明实施例提供的另一种联系人排序装置；
图6为本发明实施例提供的另一种联系人排序装置；
图7为本发明实施例提供的另一种联系人排序装置。
本发明的实施方式
下面将结合附图及具体实施例进行详细描述。
如图1所示，图1为本发明实施例提供的一种联系人排序方法流程图，包括以下步骤：
步骤S101、获取通讯录中每个联系人的情感分数。
其中，上述通讯录中每个联系人可以是通讯账号对应的通讯录中的每个联系人，上述通讯录可以用来记录联系人的信息，上述信息可以包括联系人的通讯地址、联系人的姓名、联系人的通讯账号、邮箱等等，上述通讯账号可以包括即时通讯的账号，上述每个联系人可以理解为在上述通讯账号中添加的每一个联系人，可以添加该联系人的通讯账号信息，以获得与该联系人 之间的联系。例如，用户终端通过即时通讯A的客户端注册一个账号，并在该账号的通讯录中添加了联系人以及该联系人在即时通讯A注册的账号信息。此处仅仅是举例，本实施例对此不作限定。
上述每个联系人的情感分数可以是通过通讯账号与每个联系人的通讯记录，获得的上述每个联系人的情感分数。上述通讯账号与每个联系人的通讯记录，可以是上述通讯账号与每个联系人联系时，产生的沟通记录，上述通讯记录可以是文字、图片、语音以及其他形式，上述通讯记录可以包含多种语言，可以包含多种聊天方式，如对话框聊天或者留言等等，本实施例对通讯的内容、方式不作限定。上述情感分数可以是根据上述通讯记录，判断上述通讯记录的情感极性，并计算上述通讯记录的情感分数，从而计算上述联系人的情感分数。
获取通讯录中每个联系人的情感分数可以是分别获取通讯录中每一个联系人的情感分数。
步骤S102、按照每个联系人的情感分数对所述通讯录中的联系人进行排序。
该步骤可以是将步骤S101中获取的通讯录中每个联系人的情感分数进行排序，可以是按照分数从高到低或者从低到高进行排序，也可以是依照其他的规则进行排序。依据情感分数的排序，将情感分数对应的联系人同时进行排序。例如，通讯录中有联系人A、B、C、D和E，如果联系人的情感分数排序从高到低是C、B、A、E、D，那么按照联系人的情感分数对联系人排序的结果为C、B、A、E、D。
本发明实施例可以应用于终端，例如：手机、平板电脑(Tablet Personal Computer)、膝上型电脑(Laptop Computer)、个人数字助理(personal digital assistant，简称PDA)、可穿戴式设备(Wearable Device)等。
本发明实施例通过获取通讯录中每个联系人的情感分数；按照每个联系人的情感分数对所述通讯录中的联系人进行排序，这样，能够快速找到需要查找的联系人。
如图2所示，图2为本发明实施例提供的另一种联系人排序方法流程图，所述通讯录中包括目标联系人，所述方法包括以下步骤：
步骤S201、获取所述目标联系人的通讯记录。
其中，上述目标联系人可以是通讯录中的任意一个联系人。需要说明的是，上述目标联系人可以是通讯录中包含多个联系人的联系人群组中的任一个联系人，在上述多个联系人中包含有目标联系人，对目标联系人进行情感分数计算，当然，上述通讯录中的联系人，除了上述目标联系人，其他联系人也同样执行上述情感分数计算的步骤。为了描述清楚，此处选择上述通讯录中的任一联系人即上述目标联系人，说明计算情感分数的过程。
另外，上述通讯记录可以是通讯账号与上述目标联系人的聊天记录，该聊天记录内容可以包括文字、图片、音频等形式，本实施例对此不作限定。上述获取所述目标联系人的通讯记录可以是获取通过上述通讯账号发送至上述目标联系人的通讯记录。
步骤S202、对所述通讯记录进行分词。
该步骤可以是将上述通讯记录的内容进行分词，使上述通讯记录的内容分解为多个单独的词。值得注意的是，在该步骤中，在上述通讯记录中，若采用不同的方式来表达内容，或者是采用特殊格式的内容，可以将上述内容转换为预先设定的形式。例如，采用图片、动画等方式来表达语言的，可以将上述图片、动画等内容转换为与之意思相同或相近的文本内容；对于上述通讯记录文本内容包含俚语的，可以将以俚语表达的内容转换为书面语等等。
步骤S203、计算所述分词后的文本内容的情感分数，并将该情感分数作为所述目标联系人的情感分数。
其中，上述计算所述分词后的文本内容的情感分数可以是根据步骤S202中将上述通讯记录分词后的文本内容进行情感极性判断，并计算上述分词后的文本内容的情感分数。上述计算分词后的文本内容的情感分数可以是采用经过训练的朴素贝叶斯算法进行计算，上述计算出来的情感分数可以是目标联系人的情感分数。
步骤S204、按照每个联系人的情感分数对所述通讯录中的联系人进行排序。
可选的，所述对所述通讯记录进行分词，包括：
将所述通讯记录中的通讯内容转换为满足预设条件的文本内容；
对所述文本内容进行分词。
其中，上述将上述通讯内容转换为满足预设条件的文本内容可以是将上述通讯内容转换为格式、语言、表达方式等等都满足预设的条件的文本内容，使上述通讯内容更容易被识别，且识别更准确。例如，将文本内容中采用繁体输入的文本内容的格式转换为简体格式，输入法中在全角模式下输出的文本内容的格式转换为在半角模式下输出的格式；可以将文本内容中的俚语转换为书面语，例如，文本内容中包含“杯具了”，可以转换为“悲剧了”；可以将通讯内容中采用图片等方式输入的表情或者颜文字转换为该表情或颜文字对应表达的文本内容。当然，上述转换的内容也可以是其他的情况，此处仅是举例，本实施例对此不作限定。
将上述转换后的通讯内容的文本内容进行分词。
在该实施方式中，将上述通讯记录中的通讯内容转换为满足预设条件的文本内容，对上述文本内容进行分词，可以使上述通讯内容更好地被识别，且识别更准确。
可选的，所述对所述文本内容进行分词，包括：
获取所述文本内容中能被识别的文字的分词信息；
对所述文本内容中不能被识别的文字的字向量进行网络模型运算，获得所述文本内容中不能被识别的文字的分词信息；
将所述能被识别的文字的所述分词信息和所述不能被识别的文字的所述分词信息进行维特比运算，确定所述通讯记录的文本的分词信息。
其中，上述被识别的文字可以是被识别的通用的关键字，例如，识别通讯记录内容中的人名、日期、时间、电话号码等等，上述识别可以是采用正则表达式识别上述文字，可以将上述识别的文字采用BIES分词的方法或者其他的方法，输出上述识别的文字的分词信息。例如，使用BIES分词标签的情况下，其中，“北京”可被识别，“北”和“京”的分词标签分别为“B”和“E”。此处仅是举例，本实施例对此不作限定。
上述文本内容中不能被识别的文字可以采用word2vec等工具包训练得到字向量，也可以通过随机初始化获得字向量，可以将上述字向量使用训练好的网络模型进行运算，该网络模型包括cnn卷积神经网络模型、dnn深度学习网络模型等，输出上述文本内容中不能被识别的文字的分词标签信息。
将上述文本内容中的文字的分词标签分别赋予分数，将每个字的标签分数向量合并为得分矩阵，列代表字，行代表标签，利用维特比算法进行运算，最终确定上述通讯记录的文本内容的分词信息。
在该实施方式中，将上述通讯记录中的文本内容分为能够识别和不能够识别的文字，并采用不同的方法获取文本的分词标签，能够优化资源，减少运算量，将上述能够识别和不能识别的上述文本内容运用维特比算法进行运算，得到文本内容的分词信息，使获得的上述分词信息更准确。
可选的，所述计算所述分词后的文本内容的情感分数，包括：
使用经过训练的朴素贝叶斯算法，计算所述分词后的文本内容的情感分数。
上述经过训练的朴素贝叶斯算法可以是将朴素贝叶斯算法进行训练，上述训练可以包括依据情感词典，对上述朴素贝叶斯算法进行训练，使上述朴素贝叶斯算法能够判断情感极性，并根据情感极性计算情感分数。
将上述分词信息的文本内容采用朴素贝叶斯算法进行计算，输出情感极性的分数。
本发明实施例在图1对应的实施例的基础上，通过对通讯记录转换为满足预设条件的文本内容，并将文本内容中可被识别的文本和不能被识别的文本内容分别采用不同的方式进行分词，最终确定分词信息，依据分词信息计算上述文本内容的情感分数，可以优化资源，使获取的分词信息及情感分数更准确，以使用户终端能够更快速地找到联系人。
如图3所示，图3为本发明实施例提供的一种联系人排序装置300，联系人排序装置300包括：获取模块301、排序模块302。
获取模块301，设置为获取通讯录中每个联系人的情感分数；
排序模块302，设置为按照每个联系人的情感分数对所述通讯录中的联 系人进行排序。
如图4所示，所述通讯录中至少包括目标联系人，所述获取模块301包括：
获取子模块3011，设置为获取所述目标联系人的通讯记录；
分词子模块3012，设置为对所述通讯记录进行分词；
计算子模块3013，设置为计算所述分词后的文本内容的情感分数，并将该情感分数作为所述目标联系人的情感分数。
如图5所示，所述分词子模块3012包括：
转换单元30121，设置为将所述通讯记录中的通讯内容转换为满足预设条件的文本内容；
分词单元30122，设置为对所述文本内容进行分词。
如图6所示，所述分词单元30122包括：
获取子单元301221，设置为获取所述文本内容中能被识别的文字的分词信息；
第一运算子单元301222，设置为对所述文本内容中不能被识别的文字的字向量进行网络模型运算，获得所述文本内容中不能被识别的文字的分词信息；
第二运算子单元301223，设置为将所述能被识别的文字的所述分词信息和所述不能被识别的文字的所述分词信息进行维特比运算，确定所述通讯记录的文本的分词信息。
如图7所示，所述计算子模块3013包括：
30131计算单元，设置为使用经过训练的朴素贝叶斯算法，计算所述分词后的文本内容的情感分数。
本实施例中，上述联系人排序装置300可以应用于终端，例如：手机、平板电脑(Tablet Personal Computer)、膝上型电脑(Laptop Computer)、个人数字助理(personal digital assistant，简称PDA)、可穿戴式设备(Wearable Device)等。
上述联系人排序装置300能够实现图1至图2对应的方法实施例中终端实现的各个过程，为避免重复，这里不再赘述。联系人排序装置300能够对通讯录中的联系人进行排序，使用户终端快速找到联系人。
本领域普通技术人员可以理解实现上述实施例方法的全部或者部分步骤是可以通过程序指令相关的硬件来完成，所述的程序可以存储于一计算机可读存储介质中，该程序在执行时，包括以下步骤：
获取通讯录中每个联系人的情感分数；
按照每个联系人的情感分数对所述通讯录中的联系人进行排序。
可选的，所述通讯录中至少包括目标联系人，所述获取通讯录中每个联系人的情感分数，包括：
获取所述目标联系人的通讯记录；
对所述通讯记录进行分词；
计算所述分词后的文本内容的情感分数，并将该情感分数作为所述目标联系人的情感分数。
可选的，所述对所述通讯记录进行分词，包括：
将所述通讯记录中的通讯内容转换为满足预设条件的文本内容；
对所述文本内容进行分词。
可选的，所述对所述文本内容进行分词，包括：
获取所述文本内容中能被识别的文字的分词信息；
对所述文本内容中不能被识别的文字的字向量进行网络模型运算，获得所述文本内容中不能被识别的文字的分词信息；
将所述能被识别的文字的所述分词信息和所述不能被识别的文字的所述分词信息进行维特比运算，确定所述通讯记录的文本的分词信息。
可选的，所述计算所述分词后的文本内容的情感分数，包括：
使用经过训练的朴素贝叶斯算法，计算所述分词后的文本内容的情感分数。
本领域普通技术人员可以理解，上文中所公开方法中的全部或某些步骤、系统、装置中的功能模块/单元可以被实施为软件、固件、硬件及其适当的组合。在硬件实施方式中，在以上描述中提及的功能模块/单元之间的划分不一定对应于物理单元的划分；例如，一个物理组件可以具有多个功能，或者一个功能或步骤可以由若干物理组件合作执行。某些组件或所有组件可以被实施为由处理器，如数字信号处理器或微处理器执行的软件，或者被实施为硬件，或者被实施为集成电路，如专用集成电路。这样的软件可以分布在计算机可读介质上，计算机可读介质可以包括计算机存储介质(或非暂时性介质)和通信介质(或暂时性介质)。如本领域普通技术人员公知的，术语计算机存储介质包括用于存储信息(诸如计算机可读指令、数据结构、程序模块或其他数据)的任何方法或技术中实施的易失性和非易失性、可移除和不可移除介质。计算机存储介质包括但不限于RAM、ROM、EEPROM、闪存或其他存储器技术、CD-ROM、数字多功能盘(DVD)或其他光盘存储、磁盒、磁带、磁盘存储或其他磁存储装置、或者可以用于存储期望的信息并且可以被计算机访问的任何其他的介质。此外，本领域技术人员公知的是，通信介质通常包含计算机可读指令、数据结构、程序模块或者诸如载波或其他传输机制之类的调制数据信号中的其他数据，并且可包括任何信息递送介质。。
以上所述是本发明的可选实施方式，应当指出，对于本技术领域的普通技术人员来说，在不脱离本发明所述原理的前提下，还可以作出若干改进和润饰，这些改进和润饰也应视为本发明的保护范围。
工业实用性
上述技术方案能够快速找到需要查找的联系人，简化查找程序，提高查找效率。

Claims (11)

  1. 一种联系人排序方法，包括：
    获取通讯录中每个联系人的情感分数(S101)；
    按照每个联系人的情感分数对所述通讯录中的联系人进行排序(S102)。
  2. 如权利要求1所述的方法，其中，所述通讯录中包括目标联系人，所述获取通讯录中每个联系人的情感分数(S101)，包括：
    获取所述目标联系人的通讯记录(S201)；
    对所述通讯记录进行分词(S202)；
    计算所述分词后的文本内容的情感分数，并将该情感分数作为所述目标联系人的情感分数(S203)。
  3. 如权利要求2所述的方法，其中，所述对所述通讯记录进行分词(S202)，包括：
    将所述通讯记录中的通讯内容转换为满足预设条件的文本内容；
    对所述文本内容进行分词。
  4. 如权利要求3所述的方法，其中，所述对所述文本内容进行分词，包括：
    获取所述文本内容中能被识别的文字的分词信息；
    对所述文本内容中不能被识别的文字的字向量进行网络模型运算，获得所述文本内容中不能被识别的文字的分词信息；
    将所述能被识别的文字的所述分词信息和所述不能被识别的文字的所述分词信息进行维特比运算，确定所述通讯记录的文本的分词信息。
  5. 如权利要求2所述的方法，其中，所述计算所述分词后的文本内容的情感分数(S203)，包括：
    使用经过训练的朴素贝叶斯算法，计算所述分词后的文本内容的情感分数。
  6. 一种联系人排序装置(300)，包括：
    获取模块(301)，设置为获取通讯录中每个联系人的情感分数；
    排序模块(302)，设置为按照每个联系人的情感分数对所述通讯录中的联系人进行排序。
  7. 如权利要求6所述的装置(300)，其中，所述通讯录中包括目标联系人，所述获取模块(301)包括：
    获取子模块(3011)，设置为获取所述目标联系人的通讯记录；
    分词子模块(3012)，设置为对所述通讯记录进行分词；
    计算子模块(3013)，设置为计算所述分词后的文本内容的情感分数，并将该情感分数作为所述目标联系人的情感分数。
  8. 如权利要求7所述的装置(300)，其中，所述分词子模块(3012)包括：
    转换单元(30121)，设置为将所述通讯记录中的通讯内容转换为满足预设条件的文本内容；
    分词单元(30122)，设置为对所述文本内容进行分词。
  9. 如权利要求8所述的装置(300)，其中，所述分词单元(30122)包括：
    获取子单元(301221)，设置为获取所述文本内容中能被识别的文字的分词信息；
    第一运算子单元(301222)，设置为对所述文本内容中不能被识别的文字的字向量进行网络模型运算，获得所述文本内容中不能被识别的文字的分词信息；
    第二运算子单元(301223)，设置为将所述能被识别的文字的所述分词信息和所述不能被识别的文字的所述分词信息进行维特比运算，确定所述通讯记录的文本的分词信息。
  10. 如权利要求7所述的装置(300)，其中，所述计算子模块(3013)包括：
    计算单元(30131)，设置为使用经过训练的朴素贝叶斯算法，计算所 述分词后的文本内容的情感分数。
  11. 一种计算机可读存储介质，存储有计算机可执行指令，所述计算机可执行指令被处理器执行时实现权利要求1至5中任一项所述的方法。
PCT/CN2017/090369 2016-08-12 2017-06-27 一种联系人排序方法和装置 WO2018028319A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610665349.6 2016-08-12
CN201610665349.6A CN107734123A (zh) 2016-08-12 2016-08-12 一种联系人排序方法和装置

Publications (1)

Publication Number Publication Date
WO2018028319A1 true WO2018028319A1 (zh) 2018-02-15

Family

ID=61162700

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/090369 WO2018028319A1 (zh) 2016-08-12 2017-06-27 一种联系人排序方法和装置

Country Status (2)

Country Link
CN (1) CN107734123A (zh)
WO (1) WO2018028319A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110377881A (zh) * 2019-06-11 2019-10-25 阿里巴巴集团控股有限公司 文本处理服务的集成方法、装置以及系统

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108494669A (zh) * 2018-04-03 2018-09-04 南京阿凡达机器人科技有限公司 一种向用户推荐社交对象的方法及系统
CN112019675A (zh) * 2020-07-15 2020-12-01 北京淇瑀信息科技有限公司 通讯录联系人的排序方法、装置及电子设备
CN112286967A (zh) * 2020-10-23 2021-01-29 上海淇玥信息技术有限公司 一种基于联系人执行业务任务的方法、装置和电子设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101494688A (zh) * 2008-01-22 2009-07-29 深圳富泰宏精密工业有限公司 通讯录联系人排序系统及方法
CN103873630A (zh) * 2014-02-26 2014-06-18 华为技术有限公司 联系人信息管理方法及装置
CN103905622A (zh) * 2012-12-27 2014-07-02 上海鼎为电子科技（集团）有限公司 移动终端中联系人的区分方法、装置及移动终端
CN104184866A (zh) * 2013-05-21 2014-12-03 中兴通讯股份有限公司 一种通讯录的联系人排序方法和系统
CN104333501A (zh) * 2013-07-22 2015-02-04 腾讯科技（深圳）有限公司 一种联系人列表的显示方法及装置
CN104917869A (zh) * 2014-03-10 2015-09-16 北京三星通信技术研究有限公司 用于移动终端的联系人信息处理方法和装置以及移动终端

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101494688A (zh) * 2008-01-22 2009-07-29 深圳富泰宏精密工业有限公司 通讯录联系人排序系统及方法
CN103905622A (zh) * 2012-12-27 2014-07-02 上海鼎为电子科技（集团）有限公司 移动终端中联系人的区分方法、装置及移动终端
CN104184866A (zh) * 2013-05-21 2014-12-03 中兴通讯股份有限公司 一种通讯录的联系人排序方法和系统
CN104333501A (zh) * 2013-07-22 2015-02-04 腾讯科技（深圳）有限公司 一种联系人列表的显示方法及装置
CN103873630A (zh) * 2014-02-26 2014-06-18 华为技术有限公司 联系人信息管理方法及装置
CN104917869A (zh) * 2014-03-10 2015-09-16 北京三星通信技术研究有限公司 用于移动终端的联系人信息处理方法和装置以及移动终端

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110377881A (zh) * 2019-06-11 2019-10-25 阿里巴巴集团控股有限公司 文本处理服务的集成方法、装置以及系统
CN110377881B (zh) * 2019-06-11 2023-04-07 创新先进技术有限公司 文本处理服务的集成方法、装置以及系统

Also Published As

Publication number Publication date
CN107734123A (zh) 2018-02-23

Similar Documents

Publication Publication Date Title
US20210216723A1 (en) Classification model training method, classification method, device, and medium
US11403680B2 (en) Method, apparatus for evaluating review, device and storage medium
US11762865B2 (en) Automatically augmenting message exchange threads based on tone of message
US10192545B2 (en) Language modeling based on spoken and unspeakable corpuses
US9740677B2 (en) Methods and systems for analyzing communication situation based on dialogue act information
US10853576B2 (en) Efficient and accurate named entity recognition method and apparatus
US10831796B2 (en) Tone optimization for digital content
US9792279B2 (en) Methods and systems for analyzing communication situation based on emotion information
US10056083B2 (en) Method and system for processing multimedia content to dynamically generate text transcript
WO2018028319A1 (zh) 一种联系人排序方法和装置
CN110321537B (zh) 一种文案生成方法和装置
US9792909B2 (en) Methods and systems for recommending dialogue sticker based on similar situation detection
WO2022001888A1 (zh) 基于词向量生成模型的信息生成方法和装置
WO2015021937A1 (zh) 用户推荐方法和装置
CN111639162A (zh) 信息交互方法和装置、电子设备和存储介质
CN112906380A (zh) 文本中角色的识别方法、装置、可读介质和电子设备
CN113051380A (zh) 信息生成方法、装置、电子设备和存储介质
CN112926300A (zh) 图像搜索方法、图像搜索装置及终端设备
WO2023129255A1 (en) Intelligent character correction and search in documents
US9251141B1 (en) Entity identification model training
CN111310065A (zh) 一种社交推荐方法、装置、服务器及存储介质
CN114298007A (zh) 一种文本相似度确定方法、装置、设备及介质
CN111143515B (zh) 文本匹配方法及装置
CN112822506A (zh) 用于分析视频流的方法和装置
CN111555960A (zh) 信息生成的方法

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: 17838449

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: 17838449

Country of ref document: EP

Kind code of ref document: A1