CN111726285B

CN111726285B - Instant messaging method and device

Info

Publication number: CN111726285B
Application number: CN202010850870.3A
Authority: CN
Inventors: 赖伟彬
Original assignee: Alipay Hangzhou Information Technology Co Ltd
Current assignee: Alipay Hangzhou Information Technology Co Ltd
Priority date: 2020-08-21
Filing date: 2020-08-21
Publication date: 2021-05-07
Anticipated expiration: 2040-08-21
Also published as: CN111726285A

Abstract

The specification discloses an instant messaging method and an instant messaging device. The method comprises the following steps: responding to an acquisition instruction of the group session unread message, and judging whether the unread message is matched with an analysis condition; if so, analyzing the unread message to obtain a plurality of semantic information of the unread message; and returning the semantic information to the client so that the client can display the semantic information in the group session window.

Description

Instant messaging method and device

Technical Field

The present disclosure relates to the field of internet technologies, and in particular, to an instant messaging method and an instant messaging device.

Background

Instant messaging software has become an important medium for daily communication of people, and how to assist users in improving the efficiency of communication is always a key point of attention in the industry.

Disclosure of Invention

In view of the above, the present specification provides an instant messaging method and an instant messaging device.

Specifically, the description is realized by the following technical scheme:

according to a first aspect of the present application, an instant messaging method is provided, which is applied to a server and includes:

responding to an acquisition instruction of the group session unread message, and judging whether the unread message is matched with an analysis condition;

if so, analyzing the unread message to obtain a plurality of semantic information of the unread message;

and returning the semantic information to the client so that the client can display the semantic information in the group session window.

According to a second aspect of the present application, there is provided an instant messaging method applied to a client, including:

responding to an opening instruction of a group session window, sending a group session unread message acquisition instruction to a server, so that the server analyzes and obtains a plurality of semantic information of the unread message when determining that the unread message is matched with an analysis condition;

receiving the semantic information returned by the server;

displaying the semantic information in the group session window.

According to a third aspect of the present application, there is provided an instant messaging device applied to a server, comprising:

the condition matching unit is used for responding to an acquisition instruction of the group session unread message and judging whether the unread message is matched with an analysis condition;

the message analysis unit is used for analyzing the unread messages under the condition that the unread messages are matched with analysis conditions to obtain a plurality of semantic information of the unread messages;

and the information sending unit is used for returning the semantic information to the client so that the client can display the semantic information in the group session window.

According to a fourth aspect of the present application, there is provided an instant messaging device applied to a client, including:

the instruction sending unit is used for responding to an opening instruction of a group session window and sending a group session unread message acquisition instruction to the server so that the server analyzes and obtains a plurality of semantic information of the unread message when the server determines that the unread message is matched with an analysis condition;

the information receiving unit is used for receiving the semantic information returned by the server;

the first information display unit is used for displaying the semantic information in the group session window.

According to a fifth aspect of the present application, there is provided an instant messaging device comprising:

a processor;

a memory for storing machine executable instructions;

wherein the processor is caused to:

One embodiment of the present specification realizes that, when the unread message matches the parsing condition, the server parses the unread message to obtain the semantic information of the unread message, and returns the semantic information to the client for display. Through the semantic information, the user can quickly acquire the important content of the unread messages without checking the unread messages one by one, and the communication efficiency is improved.

Drawings

Fig. 1 is a schematic diagram of an architecture of a communication system according to an exemplary embodiment of the present disclosure.

Fig. 2 is a flowchart illustrating an instant messaging method according to an exemplary embodiment of the present disclosure.

Fig. 3 is a flowchart illustrating another instant messaging method according to an exemplary embodiment of the present disclosure.

Fig. 4 is a flowchart illustrating another instant messaging method according to an exemplary embodiment of the present disclosure.

Fig. 5 is a flowchart illustrating another instant messaging method according to an exemplary embodiment of the present disclosure.

Fig. 6 is a flowchart illustrating another instant messaging method according to an exemplary embodiment of the present disclosure.

Fig. 7 is a schematic diagram illustrating a client presentation page according to an exemplary embodiment of the present specification.

Fig. 8 is a schematic diagram of another client presentation page shown in an exemplary embodiment of the present specification.

Fig. 9 is a schematic diagram of another client presentation page shown in an exemplary embodiment of the present specification.

Fig. 10 is a schematic diagram of another client presentation page shown in an exemplary embodiment of the present specification.

Fig. 11 is a schematic diagram of another client presentation page shown in an exemplary embodiment of the present specification.

Fig. 12 is a schematic structural diagram of an instant messaging device according to an exemplary embodiment of the present disclosure.

Fig. 13 is a block diagram of an instant messaging device shown in an exemplary embodiment of the present description.

Fig. 14 is a schematic structural diagram of an instant messaging device according to an exemplary embodiment of the present disclosure.

Fig. 15 is a block diagram of an instant messaging device shown in an exemplary embodiment of the present description.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present specification. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the specification, as detailed in the appended claims.

The terminology used in the description herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the description. As used in this specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

It should be understood that although the terms first, second, third, etc. may be used herein to describe various information, these information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, the first information may also be referred to as second information, and similarly, the second information may also be referred to as first information, without departing from the scope of the present specification. The word "if" as used herein may be interpreted as "at … …" or "when … …" or "in response to a determination", depending on the context.

The instant messaging software can support two-person conversation and also can support group conversation of multiple persons. For a group conversation, a plurality of users can chat together, which often generates a large amount of chat messages, and if the users fail to read the chat messages in time, the users may accumulate unread messages, for example, several tens, hundreds, or even more unread messages. For the user, in the face of a large number of unread messages, reading one by one not only takes much time, but also some key information may be missed.

The embodiment of the specification provides an instant messaging method. In one example, the method may be applied in a communication system as shown in fig. 1.

Fig. 1 is a schematic diagram of an architecture of a communication system according to an exemplary embodiment of the present disclosure. As shown in fig. 1, the system may include a network 102, a server 104, an electronic device 106, an electronic device 108, and the like.

The electronic devices 106 and 108 may be mobile devices such as a mobile phone and a palm computer, or other electronic devices such as a PC, which are only exemplary and not limiting, and the electronic devices may run a client of the instant messaging software.

It should be noted that: a client of the instant messaging software can be pre-installed on the electronic device so that the client can be started and run on the electronic device; of course, when an online "client" such as HTML5 technology is employed, the client is obtained and run without installation on the electronic device.

The server 104 may be a physical server comprising a separate host, or the server 104 may be a virtual server carried by a cluster of hosts, and the server 104 may run a server-side program of the instant messaging software.

Network 102 may include various types of wired or wireless networks. For example, the Network 102 may include the Public Switched Telephone Network (PSTN) and the Internet. Meanwhile, communication interaction can be performed between electronic devices and between the electronic devices and the server through the network 102.

Fig. 2 is a flowchart illustrating an instant messaging method according to an exemplary embodiment of the present disclosure, which may be applied to a server, such as the server 104 in fig. 1. As shown in fig. 2, the method comprises the steps of:

step S202: and responding to an acquisition instruction of the group session unread message, and judging whether the unread message is matched with a resolution condition.

In this embodiment, when the user opens the group session window, the client may send an instruction for acquiring the group session unread message to the server, and generally, the instruction may include a group identifier and a user identifier.

The server can interact with the client through the network to obtain the instruction sent by the client. When an acquisition instruction of a group session unread message sent by a client is received, a server firstly determines a corresponding group through a group identifier in the instruction, determines a corresponding user through a user identifier in the instruction, then searches the group and the unread message corresponding to the user, and then judges whether the unread message is matched with an analysis condition according to a preset analysis condition. If yes, executing step S204; if not, the found unread message can be directly sent to the client.

Step S204: and if the semantic information of the unread message is matched with the semantic information of the unread message, analyzing the unread message to obtain a plurality of semantic information of the unread message.

The server can analyze the unread message according to a preset method to obtain corresponding semantic information.

In one embodiment, the server may parse the unread message based on a keyword parsing model to obtain a number of keywords of the unread message. In another embodiment, the server may parse the unread message based on a digest generation model to obtain several digests of the unread message. Or, the server may also use the methods of the two embodiments to generate the keywords of the unread message and generate the semantic information of the unread message.

Step S206: and returning the semantic information to the client so that the client can display the semantic information in the group session window.

Based on the semantic information generated in step S204, the server may send the semantic information to the client, so that the client displays the semantic information to the user in the group session window.

It will be appreciated that the server will also typically send unread messages to the client.

At this point, the flow shown in fig. 2 is completed. By the method, the client can show the semantic information of the unread messages to the user, so that the user can quickly acquire the important contents of the unread messages through the semantic information without checking the unread messages one by one, and the communication efficiency is improved.

Fig. 3 is a flowchart illustrating another instant messaging method that may be applied to a client, such as the electronic device 106 in fig. 1, according to an exemplary embodiment of the present disclosure. As shown in fig. 3, the method comprises the steps of:

step S302: and responding to an opening instruction of the group session window, and sending a group session unread message acquisition instruction to the server so that the server analyzes and obtains a plurality of semantic information of the unread message when the server determines that the unread message is matched with an analysis condition.

In this embodiment, the client may present the contact information, such as a group name, a group avatar, and the like, in the contact list. The user may issue an instruction to open the group session window through the contact list displayed by the client, for example, the user may click on the group name.

After receiving an opening instruction of a group session window issued by a user, the client may send an unread group session message acquisition instruction for the group to the server. The obtaining instruction may carry a group identifier and a user identifier.

After receiving the obtaining instruction, the server may obtain a plurality of semantic information of the corresponding unread message through parsing according to the flow shown in fig. 2.

Step S304: and receiving the semantic information returned by the server.

Step S306: displaying the semantic information in the group session window.

In this embodiment, the client may receive an unread message returned by the server.

If the server determines that the unread message is matched with the preset analysis condition, the client can also receive semantic information sent by the server.

And according to the received information, the client displays the unread message and the semantic information to the user through the group session window.

At this point, the flow shown in fig. 3 is completed. Through the method, the client can show the semantic information of the unread message to the user. The user can quickly acquire the important content of the unread message through the semantic information without checking the unread message one by one, so that the communication efficiency is improved.

One embodiment of the method of the present disclosure is described below in conjunction with fig. 4.

Fig. 4 is a flowchart illustrating another instant messaging method according to an exemplary embodiment of the present disclosure, which may be applied to the communication system shown in fig. 1, as shown in fig. 4. Wherein, the client in the method can be operated in the electronic device of fig. 1, such as the electronic device 106; the server in the method may operate in the server 104 in fig. 1. The method may comprise the steps of:

step S402: and the client responds to the opening instruction of the group session window and sends a group session unread message acquisition instruction to the server.

In this embodiment, the user may open the group session window in a variety of ways. For example, the user may click on the group id with a mouse, or may open the group session window with voice control. Of course, in practical applications, other ways may also be used to implement the opening of the group session window by the user, which is only an exemplary way and is not particularly limited.

In this embodiment, in response to the opening instruction of the group session window, the client may send a group session unread message acquisition instruction to the server. The client may send a group session unread message acquisition instruction to the server through the network, where the instruction carries a group identifier, such as a group ID.

Step S404: and the server side responds to the acquisition instruction of the group session unread message and judges whether the unread message is matched with the analysis condition.

In this embodiment, after receiving an instruction for acquiring a group session unread message sent by a client, a server may query a corresponding unread message according to a group identifier carried in the instruction, match the unread message with a preset parsing condition, and if the parsing condition is matched, execute step S406.

If the parsing condition is not matched, the server side can directly send the inquired unread message to the client side.

In one example, the server may determine whether the parsing condition is matched according to the number of unread messages. The server side can count the number of the unread messages and judge whether the number of the unread messages is larger than a preset number threshold. Wherein, the number threshold may be uniformly set for all groups, for example, the same setting is 20; or may be in one-to-one correspondence with each group, for example, the number threshold set by the user for each group may be received by the client, for example, the number threshold is set to 20 for group 1, the number threshold is set to 100 for group 2, and the like. And are not limited herein.

If the number of the unread messages is greater than the preset number threshold, it indicates that there are many unread messages currently, and if the unread messages are directly displayed to the user, the user may need to spend more time to read, which is low in efficiency. The server may determine that the unread message matches the parsing condition, and perform step S406 to parse the unread message.

If the number of the unread messages is not greater than the preset number threshold, the number of the unread messages is less, and the time consumed by the user for reading the unread messages is less. The server side does not need to analyze the unread messages, and directly sends the unread messages to the client side for display.

For example, if the parsing condition of the group 1 is that the number of unread messages is greater than 20, the server may determine whether the number of unread messages corresponding to the user is greater than 20. If the number of the unread messages of the group 1 by the user 1 is greater than 20 (for example, 21), the server may determine that the unread messages match the parsing condition, and execute step S406 to parse the unread messages; if the number of the unread messages of the group 1 by the user 1 is less than or equal to 20 (for example, 10), the server may determine that the unread messages do not match the parsing condition, and directly send the unread messages to the client.

In another example, the server may determine whether the parsing condition is matched according to the sending time span of the unread message. The embodiment is suitable for the situation that the number of the unread messages is greater than 1, the server may first determine the earliest sending time and the latest sending time of the unread messages, where the earliest sending time is the sending time with the longest time difference from the current time, the latest sending time is the sending time closest to the current time, and then the server takes the time difference between the latest sending time and the earliest sending time as a sending time span and determines whether the sending time span is greater than a preset time threshold.

Wherein, the time threshold may be uniformly set for all groups, for example, uniformly set to 12 hours; or may be in one-to-one correspondence with each group, for example, the time threshold set by the user for each group may be received by the client, for example, the time threshold set for group 1 is 12 hours, the time threshold set for group 2 is 2 hours, and the like. And are not limited herein.

If the sending time span is greater than the preset time threshold, it indicates that the unread message is longer than the current time, the value of the unread message to the user may not be high, and the user does not need to read the unread message one by one, and the server may determine that the unread message matches the parsing condition, and execute step S406 to parse the unread message.

If the number of the unread messages is not larger than the preset time threshold, the unread messages are close to the current time, and the value of the unread messages to the user is probably high. The server side does not need to analyze the unread messages, and directly sends the unread messages to the client side for display.

For example, if the parsing condition of the group 1 is that the sending time span of the unread message is greater than 2 hours, the server may determine whether the sending time span of the unread message corresponding to the user is greater than 2 hours. If the sending time span of the unread message of the group 1 by the user 1 is greater than 2 hours (for example, 5 hours), the server may determine that the unread message matches the parsing condition, and execute step S406 to parse the unread message; if the sending time span of the unread message of the group 1 by the user 1 is less than or equal to 2 hours (for example, 10 minutes), the server may determine that the unread message does not match the parsing condition, and directly send the unread message to the client.

Of course, the server may also combine two analysis conditions, i.e., the number of unread messages and the sending time span, to determine the number of unread messages simultaneously, or may also set other analysis conditions, which is not limited herein.

Step S406: and if the unread message is matched with the analysis condition, the server analyzes the unread message to obtain a plurality of semantic information of the unread message.

In this embodiment, based on the matching result in step S204, if the unread message matches the parsing condition, the server analyzes the unread message. For example, the server may parse the unread message according to the flow shown in fig. 5.

As shown in fig. 5, the method comprises the following steps:

step S502: and combining the unread messages sent by the same sending member to obtain a plurality of unread message sets.

In instant messaging software, a message sent by a user is usually a short spoken sentence, and a complete sentence may be split into a plurality of phrases to be sent respectively. For example, user a may send the following 4 messages: "classmates, praise", "hurry congregation", "401 meeting room", "desperate money and village, west". It will be appreciated that the plurality of messages sent by the user a are actually described in terms of the same subject.

In order to be closer to the use habit of the user, the server side can combine the unread messages sent by the same sending member. For example, unread messages sent by the same sending member and sent adjacent in time can be combined into one set of unread messages. Here, the transmission time adjacency means that the transmission time is adjacent in all the history session messages, that is, a plurality of unread messages are consecutive and the plurality of unread messages are transmitted by the same member.

In implementation, the server may obtain all unread messages and arrange the unread messages into a sequence from early to late according to the sending time. For each unread message in the sequence, the server may determine whether the unread message and the next unread message are sent by the same sending member, and if so, may divide the unread message and the next unread message into the same unread set.

For example, for the first unread message m1, it is first determined that the sending member of the unread message m1 is user u1, and then it is determined whether the sending member of the second unread message m2 in the sequence is user u 1. If not, the merging process of m1 is ended, and m2 is merged according to a similar method; if yes, whether the sending member of the third unread message m3 is the user u1 … … is continuously judged until the sending member is not the N +1 th unread message m (N + 1) of u1, and accordingly, the first unread message m1 to the N th unread message mN can be merged into an unread message set. Other unread messages can also be merged according to the same method, which is not described herein again.

The method for merging unread messages in this embodiment is specifically described below by taking table 1 as an example.

As shown in table 1, table 1 includes 5 unread messages sorted from morning to evening by the sending time.

In this embodiment, 5 pieces of unread messages shown in table 1 may be merged, and for the 1 st unread message "classmates, a huila", the unread messages that are the same as the sending member of the unread message and whose sending time is adjacent to each other are sequentially found to include "401 meeting room" and "desquay money bank, west".

Thus, the 3 unread messages can be combined into one set of unread messages, for example, can be combined into { "classmates, funeral, 401 conference room", "desperate money house, west" }. Alternatively, the sending member information may also be added to the unread message set, and may be merged into, for example, { "sender: user a "," students, a praise "," 401 meeting room "," desperate money house, west "}.

Then, the 4 th unread message "received, coming right now" is merged. Obviously, there is no unread message in table 1 that is the same as the sending member of the unread message and has a sending time adjacent, so the unread message can be separately formed into a set, for example { "sender: user B "," receive, go right now ".

Continuing then with the merge of the 5 th unread message "other classmates, quick hits", the unread messages can be formed separately into a set, as such, for example { "sender: user a "," other classmates, a quick action "}.

At this point, the merging of the 5 unread messages shown in Table 1 is complete.

In another embodiment, the merging may also be based on the transmission time span of the unread messages. For example, a smaller time interval threshold (e.g., 10 seconds) may be set, and then multiple unread messages sent by the same sending member and having a sending time span smaller than the time interval threshold may be merged. And will not be described in detail herein.

Based on the merging operation of the unread messages sent by the same sending member, this embodiment may obtain a plurality of unread message sets, where each unread message set includes at least one unread message.

Step S504: and analyzing the semantic information of each unread message set.

In this embodiment, the semantic information may include a plurality of categories, and the common categories are related to keywords and abstracts. The following is illustrated by 2 examples respectively:

example 1: the server side obtains a plurality of keywords as semantic information of the unread messages based on the unread message set analysis.

In this embodiment, a keyword parsing model may be used to parse the unread message set. As shown in fig. 6, the following steps may be included:

step S5042: and aiming at each unread message set, adopting a plurality of keyword analysis models to analyze the keywords of the unread message set to obtain a plurality of keyword sets corresponding to the keyword analysis models.

In one example, before the text message is parsed by using the keyword parsing model, each unread message in the unread message set may be segmented, wherein the segmentation algorithm may refer to a related technology, such as a reverse maximum Matching algorithm (RMM), a Bi-directional maximum Matching method (BM), and the like; then preprocessing the word segmentation result, such as word cleaning and the like, and filtering out words with low equivalence values, such as 'good' and 'possible' when the words are cleaned; and then inputting the preprocessed words into a keyword analysis model for analysis.

In another example, the unread message set may be directly input into the keyword parsing model, and the keyword parsing model performs word segmentation and preprocessing, and then parses the preprocessed words.

The method for analyzing the unread message set after word segmentation processing by using the keyword model is described below.

In this embodiment, for each unread message set after word segmentation, at least one keyword analysis model may be used to analyze the unread message set, and each keyword analysis model may output one keyword set. Based on this, for each unread message set, a plurality of keyword analysis models can be adopted to obtain a plurality of keyword sets corresponding to the keyword analysis models.

For example, assuming that the unread message set M is { "students, praise", "401 conference room", "desperate money house, west" }, the keyword parsing model 1 and the keyword parsing model 2 may perform word segmentation processing and parsing on the unread message set, respectively, and the output keyword set may include keywords and confidence degrees corresponding to the keywords, as shown in table 2.

Step S5044: and synthesizing each keyword set to obtain the target keywords of the unread message set.

In this embodiment, for each unread message set, a score of each keyword may be determined according to the keywords and the confidence degrees in several keyword sets of the unread message set, and then at least one target keyword is determined based on the scores of the keywords.

In one example, the user presets a weight for each keyword parsing model in advance. In this embodiment, first, the weight set by the user for the keyword analysis model is used as the weight of the corresponding keyword set; then, for each keyword, the confidence weighted sum of the keyword in each keyword set can be used as the score of the keyword; and then determining the target keywords of the unread message according to the scores of the keywords.

Taking the keyword set in table 2 as an example, assuming that the weights set by the keyword analysis model 1 and the keyword analysis model 2 are 0.4 and 0.6, respectively, it can be determined that the score corresponding to each keyword is as shown in table 3:

in table 3, a model 1 weight score represents a product of a confidence of each keyword in the keyword analysis model 1 and a weight corresponding to the keyword analysis model 1, and a model 2 weight score represents a product of a confidence of each keyword in the keyword analysis model 2 and a weight corresponding to the keyword analysis model 2, and a score of each keyword can be determined by integrating the weight scores of the two models.

In another example, the sum of the confidence levels of the keywords in each model may be directly used as the score for the keyword. Alternatively, other methods may be used to determine the score for each keyword, and are not limited herein.

In this embodiment, a score threshold may be preset, and then a specified number of target keywords may be screened out from the keywords with scores higher than the score threshold, for example, the keywords with scores higher than the score threshold may be screened out, then the keywords may be ranked in order of scores from high to low, and the keywords ranked as top N may be used as the target keywords.

For example, for the keyword scores shown in table 3, assuming that the score threshold is 0.15, the keywords having scores higher than the threshold are first screened out, and then sorted from high to low according to the scores as "401 meeting room", "deshe", "meetings", "money", and then selected according to the set number of target keywords. For example, if the target keyword is 4, the 4 keywords may be used as the target keyword; if the number of the target keywords is set to 3, since there are two keywords ("conference" and "money") ranked at 3 rd, one keyword can be randomly selected from them.

Of course, the method for selecting the target keyword is only described as an example, and other methods may be used for selection in practical applications, which is not specifically limited in this embodiment.

In this embodiment, after the target keywords are screened out, the target keywords may also be merged. In implementation, for each target keyword, whether other target keywords are adjacent to the target keyword in the original unread message or not can be judged, and if so, merging operation can be performed.

For example, in table 3, the two keywords "desperate" and "qianzhuang" are adjacent in the original unread message "desperate qianzhuang, west", and then they can be merged into "desperate qianzhuang".

Of course, in other examples, the keywords may also be merged before the target keywords are screened, for example, whether any two keywords are adjacent in the original unread message is determined, if so, the two keywords may be merged, and the confidence level of the two keywords is predicted by using the keyword analysis model after merging. Or, the keywords after the confidence degree is predicted may also be merged, and the score of the merged keywords may be the average score of the two keywords, and the like, which is not limited in this specification.

Step S5046: determining semantic information of the set of unread messages based on the target keywords.

In this embodiment, for each unread message set, semantic information of the unread message set may be determined according to its target keywords.

In one example, the target keyword determined in step S5044 may be directly used as the semantic information of the unread message set. For example, in the example shown in table 3 above, the semantic information may be determined as 4 keywords: "401 meeting room", "desperate", "money bank", "conference".

Of course, in other examples, the keywords may also be merged before the target keywords are screened, for example, whether any two keywords are adjacent in the original unread message is determined, if so, the two keywords may be merged, and the confidence level of the two keywords is predicted by using the keyword analysis model after merging. Alternatively, the keywords after the confidence degree is predicted may also be merged, and the score of the merged keyword may be the average value or the maximum value of the scores of the two keywords, which is not limited in this specification. Thus, the description of embodiment 1 is completed.

In the embodiment, the server generates the keywords as the semantic information for the unread messages, and returns the semantic information to the client and displays the semantic information to the user, so that the user can quickly know the key information in the unread messages according to the keywords, and the reading efficiency of the user is improved.

On the other hand, the embodiment analyzes the unread message by adopting a plurality of keyword analysis models, so that the influence of inaccuracy of a single model on the overall analysis effect can be avoided. In addition, in this embodiment, weights may be set for the plurality of keyword analysis models, and the output result and the weights may be considered in combination, so that the accuracy of the analysis result may be improved.

Of course, other methods may be used to analyze the keywords of the unread message, for example, 1 keyword analysis model may be used to analyze the keywords of the unread message set. This is by way of example only and not by way of limitation.

Example 2: and the server side analyzes the unread message set to obtain a plurality of abstracts as semantic information of the unread message.

In this embodiment, a digest generation model may be used to parse the unread message set.

Generally, the server may be preset with a trained abstract generation model. For example, the summary generation model may be trained based on the Seq2Seq algorithm using the history chat record as the sample data.

In one example, the set of unread messages includes at least one message text. And inputting the unread message set into the digest generation model to obtain the digest of the unread message. For example, for the unread message set { "classmates, kela", "401 conference room", "desquay chien, west" } shown in embodiment 1, the abstract can be parsed to be "desquay chien 401 conference room meeting".

In one example, the set of unread messages input into the digest generation model may also include the sending member of the message. For example, assuming that the sender of the unread message set M is user a, the unread message set may be { "sender: user a "," students, a praise "," 401 meeting room "," desperate money house, west "}. For the set of unread messages, the summary can be parsed to "user a notifies students to meet in the conference room of the desperate money bank 401".

Of course, the digest generation model may also output a plurality of digests for one set of unread messages. For example, assume that the set of unread messages is { "sender: user a, students, a party, a 401 meeting room, a desperate money bank, a west, and a B-band notebook, the parsed summary can be "user a notifies the students to the desperate money bank 401 meeting room and" user a notifies the B-band notebook ".

This completes the description of example 2. In the embodiment, the server generates the abstract for the unread message as the semantic information, and returns the semantic information to the client and displays the semantic information to the user, so that the user can quickly know the approximate content in the unread message according to the abstract, and the reading efficiency of the user is improved.

Of course, in this embodiment, the keyword or the abstract is used as the semantic information, or in this embodiment, the keyword and the abstract may be output together as the semantic information by using the methods shown in the above two embodiments. And are not limited herein.

Step S506: and synthesizing the semantic information of each unread message set to obtain the semantic information of the unread messages of the group conversation.

The foregoing steps may determine that the semantic information of the unread message set includes a number of keywords and/or a number of digests. In this embodiment, the semantic information of the unread messages may be summarized according to a preset method, for example, if the semantic information output by a plurality of unread message sets is repeated, the repeated semantic information may be deleted.

In one example, semantic information may be filtered to filter out semantic information with less meaning. For example, if the semantic information of the unread message set includes a keyword such as "received", the keyword may be deleted directly.

Of course, this is merely an example, and the semantic information of the unread message sets may be summarized by other methods.

Thus, the flow shown in fig. 5 is completed, and the server uses the summarized semantic information as the semantic information of the group session unread message.

Step S408: and the server side returns the semantic information to the client side.

In this embodiment, the server may return the queried unread message and the semantic information obtained in step S406 to the client.

Optionally, the server may further establish a corresponding relationship between the unread message and the semantic information, and return the corresponding relationship to the client.

Step S410: and the client receives the semantic information returned by the server.

Step S412: and the client displays the semantic information in the group session window.

In this embodiment, the group conversation window may include a conversation message display area and a message editing area.

As shown in fig. 7, the group conversation window 700 includes a message display area 702 and a message editing area 704. Wherein the conversation message presentation area 702 can present historical conversation messages to the user. The message edit area 704 may show the user the text or picture, etc. the user is entering.

Optionally, the group session window 700 may further include: a navigation bar 706 for displaying reading progress; a group member display area 708 for displaying group members; a group name display area 710 for displaying the group name and the like.

In this embodiment, the client may present the unread message and the semantic information received from the server in the message presentation area. As shown in fig. 8, in the group session window 700, the message presentation area 702 may include two sub-areas, which are an unread message presentation sub-area 7022 and a semantic information presentation sub-area 7024, respectively. The unread message display sub-region 7022 is configured to display the unread message; the semantic information presentation sub-region 7024 is used to present semantic information. Optionally, the unread message presentation sub-region 7022 can be located above the semantic information presentation sub-region 7024.

In implementation, the client will generally determine the display proportion of the semantic information display sub-region 7024 in the message display region 702 according to the amount of the semantic information; then, the display proportion of the unread message display sub-region 7022 in the message display region 702 is determined according to the display proportion, and the unread message is displayed in the unread message display sub-region 7022.

Optionally, after determining the display proportion of the unread messages in the unread messages display sub-region 7022, the client may determine the number of the unread messages that can be accommodated, and select a plurality of unread messages whose sending time is closest to the current time according to the number for display.

Of course, this is merely an exemplary illustration, and the client may also present the semantic information in another area other than the message presentation area 702, for example, the semantic information may be presented in the message editing area 704, and the present embodiment is not limited thereto.

In this embodiment, when the client displays the semantic information, the semantic information may be displayed in a form of a text box, and different semantic information is displayed in different text boxes. Still taking fig. 8 as an example, the semantic information presentation sub-area 7024 includes 2 pieces of semantic information, respectively, "a notify desperate money house of desperate" and "D several minutes later", which are presented in two text boxes.

In order to better distinguish the semantic information, the client can set at least one different display attribute for different semantic information when displaying the semantic information. The display attribute may include a display position, a text box format, a font, a color, and the like. For example, 2 pieces of semantic information in fig. 8, the line style of the text box corresponding to "a notifies the desquamation of the debt and money house" is in the form of "dot-line", and the line style of the text box corresponding to "D a few minutes later" is in the form of "dashed line".

In one example, the semantic information may include a variety of categories, which may include keywords and summaries, for example. The embodiment can determine different presentation attributes for each category of semantic information in the session message presentation area.

As shown in fig. 9, in the group conversation window 700, the semantic information displayed in the message display area 702 includes 2 categories, which are a summary and a keyword. The text box styles corresponding to the two semantic information are different, wherein the text box style corresponding to the abstract is a non-straight line, and the text box style corresponding to the keyword is a straight line.

In addition, the embodiment may also label a specific category of the semantic information. For example, in fig. 9, for semantic information of a digest category, a "digest" typeface is marked above a corresponding digest; for semantic information of the keyword category, a keyword word pattern is marked above the corresponding keyword.

In this embodiment, different display attributes may also be determined for semantic information of the same category. For example, for the keywords in fig. 9, the filling pattern of the text box corresponding to each keyword is different, for example, the filling pattern of the keyword "meeting" is "oblique line", the filling pattern of the keyword "set" is "grid line", and the filling pattern of the keyword "degintended money bank" is "oblique line".

Of course, the above-mentioned display attributes may be of other types, and are only exemplary and not limiting.

In this embodiment, in order to enable the user to read the unread message more conveniently, the client may further perform unread message skipping based on the semantic information, and when receiving the selection instruction for the target semantic information, the user may display the unread message corresponding to the target semantic information in the session message display area.

For example, the user may select one piece of semantic information in the session message display area through the selection instruction, and after receiving the selection instruction, the client may determine the unread message corresponding to the semantic information according to the correspondence between the unread message and the semantic information received from the server, and jump the unread message displayed in the session message display area to an unread message including the semantic information. Here, the selection instruction may include a mouse click, a touch operation, and the like, which is determined according to the actual situation and is not limited in this embodiment.

In one example, in order to more fully present the unread messages, after receiving the selection instruction of the user and performing the unread message skip operation, the client may hide the semantic information, or reduce the presentation ratio of the semantic information presentation sub-region in the session message presentation region, and expand the presentation ratio of the unread message presentation sub-region in the session message presentation region. Of course, this is merely an example and the present embodiment is not limited.

In this embodiment, the session message display area further includes a navigation bar, and the client may further establish a corresponding relationship between the navigation bar and each semantic information. Firstly, dividing the navigation bar into a plurality of navigation areas with corresponding quantity according to the quantity of the semantic information, and then determining corresponding semantic information for each navigation area. As shown in the group session window 700 of fig. 10, the navigation bar 706 may be divided into 3 navigation regions from top to bottom, wherein the first navigation region corresponds to the keyword "meeting", the second navigation region corresponds to the keyword "set", and the third navigation region corresponds to the keyword "desquay money bank".

In one example, in order to enable a user to more intuitively know the corresponding relationship between the navigation area and the semantic information, different display attributes may be set for the navigation area, and the display attribute of each navigation area is matched with the display attribute of the text box in which the semantic information corresponding to the navigation area is located. For example, different colors may be set for each navigation area, and the color of each navigation area is the same as the color of the text box of the corresponding semantic information; alternatively, different filling patterns may be set for each navigation area, and the filling pattern of each navigation area is the same as the text box filling pattern of the corresponding semantic information.

Still taking fig. 10 as an example, in the group session window 700 shown in fig. 10, the navigation bar 706 is divided into 3 navigation areas, and the filling patterns of the 3 navigation areas are different. According to the sequence from top to bottom, the filling pattern of the first navigation area is an upper oblique line, and the filling pattern of the text box of the keyword 'meeting' corresponding to the navigation area is also an upper oblique line; the filling pattern of the second navigation area is a grid line, and the text box filling pattern of the keyword set corresponding to the navigation area is also a grid line; the fill pattern of the third navigation area is "oblique line", and the fill pattern of the text box of the keyword "deshe money bank" corresponding to the navigation area is also "oblique line".

Of course, this is only an exemplary description, and the corresponding relationship between the navigation area and the semantic information may also be displayed by other types of display attributes, which is not limited in this embodiment.

It can be understood that, when reading a message, a user is usually used to click a navigation bar to position the message, in this embodiment, in order to be compatible with the use habit of the user, and to facilitate the user to view the message and improve the user experience, the skip of an unread message may be implemented based on the correspondence between the navigation bar and semantic information.

For example, when a click instruction of a user to a target navigation area is received, the client may search for target semantic information corresponding to the target navigation area, and then, the client may determine a target unread message corresponding to the target semantic information according to a correspondence between the semantic information and the unread message, and display the target unread message in the session message display area.

Still taking FIG. 10 as an example, assume that the user clicks on the first navigation area. After receiving the click instruction, the client may determine that the semantic information corresponding to the first navigation area is the keyword "meeting", and then the client may jump the unread message displayed in the session message display area to an unread message containing "meeting", as shown in fig. 11.

Fig. 11 shows a group conversation window 700, which includes a conversation message display area 702 and a message editing area 704. The unread message "classmates, praise" corresponds to the keyword "praise" in fig. 10.

The flow shown in fig. 4 is completed. In this embodiment, the server may parse the unread message and send semantic information of the unread message to the client. The client can show the semantic information of the message to the user, so that the user can quickly obtain the important content of the unread message without checking the unread message one by one, and the communication efficiency of the user is improved.

In addition, the client can also realize the skip function of the unread messages based on the selection operation of the semantic information and the navigation bar by the user, so that the user can quickly read the unread messages corresponding to the semantic information after checking the semantic information, the user can conveniently check the messages, and the communication efficiency of the user is improved.

Corresponding to the embodiment of the instant messaging method, the specification also provides an embodiment of an instant messaging device.

The embodiment of the instant messaging device can be applied to a server. The device embodiments may be implemented by software, or by hardware, or by a combination of hardware and software. Taking a software implementation as an example, as a logical device, the device is formed by reading corresponding computer program instructions in the nonvolatile memory into the memory for operation through the processor of the server where the device is located. From a hardware aspect, as shown in fig. 12, it is a hardware structure diagram of a server where the instant messaging device of this specification is located, except for the processor, the memory, the network interface, and the nonvolatile memory shown in fig. 12, the instant messaging device in the embodiment may also include other hardware according to an actual function of the server, which is not described again.

Referring to fig. 13, the instant messenger device 1300 can be applied to the server shown in fig. 12, and includes:

the condition matching unit 1302 is configured to respond to an obtaining instruction of the group session unread message, and determine whether the unread message matches an analysis condition.

A message parsing unit 1304, configured to parse the unread message to obtain a plurality of semantic information of the unread message, when the unread message matches a parsing condition.

An information sending unit 1306, configured to return the semantic information to the client, so that the client displays the semantic information in the group session window.

Optionally, the determining whether the unread message matches an analysis condition includes:

judging whether the quantity of the unread messages is greater than a quantity threshold value;

and if so, determining that the unread message matches the parsing condition.

Optionally, the determining whether the unread message matches a parsing condition includes:

judging whether the sending time span of the unread message is larger than a time length threshold value or not;

and if so, determining that the unread message matches the parsing condition.

Optionally, parsing the unread message to obtain a plurality of semantic information of the unread message, including:

combining unread messages sent by the same sending member to obtain a plurality of unread message sets;

analyzing the semantic information of each unread message set;

and synthesizing the semantic information of each unread message set to obtain the semantic information of the unread messages of the group conversation.

Optionally, merging the unread messages sent by the same sending member includes:

and combining the unread messages which are sent by the same sending member and are adjacent in sending time.

Optionally, parsing out semantic information of each set of unread messages includes:

aiming at each unread message set, adopting a plurality of keyword analysis models to analyze the keywords of the unread message set to obtain a plurality of keyword sets corresponding to the keyword analysis models;

synthesizing each keyword set to obtain target keywords of the unread message set;

determining semantic information of the set of unread messages based on the target keywords.

Optionally, determining semantic information of the set of unread messages based on the target keywords includes:

judging whether different target keywords are adjacent in the group session unread message or not;

and if the unread message sets are adjacent, merging the adjacent target keywords to obtain the semantic information of the unread message sets.

and aiming at each unread message set, generating the abstract of the unread message set by adopting an abstract generation model to serve as the semantic information of the unread message set.

The implementation process of the functions and actions of each unit in the above device is specifically described in the implementation process of the corresponding step in the above method, and is not described herein again.

For the device embodiments, since they substantially correspond to the method embodiments, reference may be made to the partial description of the method embodiments for relevant points. The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution in the specification. One of ordinary skill in the art can understand and implement it without inventive effort.

The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. A typical implementation device is a computer, which may take the form of a personal computer, laptop computer, cellular telephone, camera phone, smart phone, personal digital assistant, media player, navigation device, email messaging device, game console, tablet computer, wearable device, or a combination of any of these devices.

Corresponding to the embodiment of the instant messaging method, the specification also provides another embodiment of the instant messaging device.

The embodiment of the instant messaging device can be applied to the client. The device embodiments may be implemented by software, or by hardware, or by a combination of hardware and software. The software implementation is taken as an example, and as a logical device, the device is formed by reading corresponding computer program instructions in the nonvolatile memory into the memory for operation through the processor of the client where the device is located. From a hardware aspect, as shown in fig. 14, the instant messaging device in this specification is a hardware structure diagram of a client, and in addition to the processor, the memory, the network interface, and the nonvolatile memory shown in fig. 14, the instant messaging device in the embodiment may also include other hardware according to an actual function of the client, which is not described again.

Referring to fig. 15, the instant communication apparatus 1500 can be applied to the client shown in fig. 14, and includes:

the instruction sending unit 1502 is configured to send a group session unread message obtaining instruction to a server in response to an opening instruction of a group session window, so that the server obtains a plurality of semantic information of the unread message by parsing when determining that the unread message matches the parsing condition.

An information receiving unit 1504, configured to receive the semantic information returned by the server.

A first information showing unit 1506, configured to show the semantic information in the group session window.

Optionally, the apparatus further comprises:

a second information displaying unit 1508, configured to receive a selection instruction for target semantic information, and display an unread message corresponding to the target semantic information in a session message displaying area of the group session window.

Optionally, displaying the semantic information in the group session window includes:

and displaying the semantic information in a session message display area of the group session window in a text box form, wherein different semantic information is displayed in different text boxes.

Optionally, the apparatus further comprises:

a third information presentation unit 1510 configured to:

dividing navigation bars in a session message display area into a plurality of navigation areas, wherein different navigation areas correspond to different semantic information;

receiving a selection instruction aiming at a target navigation area, and determining target semantic information corresponding to the target navigation area;

and displaying the unread message corresponding to the target semantic information in a session message display area.

Optionally, the display attributes of the navigation area corresponding to the same semantic information and the text box corresponding to the semantic information are the same.

Optionally, the semantic information is divided into categories; the semantic information of different types at least has one different display attribute, and the semantic information of the same type at least has one same display attribute.

Optionally, the apparatus further comprises:

the fourth information presentation unit 1512 is configured to label the category of the semantic information in the session message presentation area.

Optionally, the category of the semantic information includes: keywords and summaries.

Corresponding to the embodiment of the instant messaging method, the present specification further provides an instant messaging device, including: a processor and a memory for storing machine executable instructions. Wherein the processor and the memory are typically interconnected by means of an internal bus. In other possible implementations, the device may also include an external interface to enable communication with other devices or components.

In this embodiment, by reading and executing machine-executable instructions stored by the memory corresponding to instant messaging logic, the processor is caused to:

and if so, determining that the unread message matches the parsing condition.

analyzing the semantic information of each unread message set;

receiving the semantic information returned by the server;

displaying the semantic information in the group session window.

Optionally, the method further comprises:

and receiving a selection instruction aiming at the target semantic information, and displaying the unread message corresponding to the target semantic information in a session message display area of the group session window.

Optionally, the method further comprises:

and marking the category of the semantic information in the session message display area.

In correspondence with the aforementioned embodiment of the instant messaging method, the present specification also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of:

and if so, determining that the unread message matches the parsing condition.

analyzing the semantic information of each unread message set;

receiving the semantic information returned by the server;

displaying the semantic information in the group session window.

Optionally, the method further comprises:

The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.

The above description is only a preferred embodiment of the present disclosure, and should not be taken as limiting the present disclosure, and any modifications, equivalents, improvements, etc. made within the spirit and principle of the present disclosure should be included in the scope of the present disclosure.

Claims

1. An instant messaging method is applied to a client and comprises the following steps:

receiving the semantic information returned by the server;

displaying the semantic information in a session message display area of the group session window in a text box form, wherein different semantic information is displayed in different text boxes;

dividing navigation bars in a session message display area into a plurality of navigation areas, wherein different navigation areas correspond to different semantic information; the navigation area corresponding to the same semantic information is the same as the display attribute of the text box corresponding to the semantic information;

2. The method of claim 1, further comprising:

3. The method of claim 1, the semantic information being divided into categories; the semantic information of different types at least has one different display attribute, and the semantic information of the same type at least has one same display attribute.

4. The method of claim 3, further comprising:

5. The method of claim 3, the categories of semantic information comprising: keywords and summaries.

6. An instant communication device applied to a client comprises:

the first information display unit is used for displaying the semantic information in a session message display area of the group session window in a text box mode, wherein different semantic information is displayed in different text boxes;

a third information presentation unit for:

7. The apparatus of claim 6, the apparatus further comprising:

and the second information display unit is used for receiving a selection instruction aiming at the target semantic information and displaying the unread message corresponding to the target semantic information in the conversation message display area of the group conversation window.

8. The apparatus of claim 6, the semantic information divided into categories; the semantic information of different types at least has one different display attribute, and the semantic information of the same type at least has one same display attribute.

9. The apparatus of claim 8, the apparatus further comprising:

and the fourth information display unit is used for marking the category of the semantic information in the session message display area.

10. The apparatus of claim 8, the categories of semantic information comprising: keywords and summaries.

11. An instant messaging device comprising:

a processor;

a memory for storing machine executable instructions;

wherein the processor is caused to:

receiving the semantic information returned by the server;