CN110391973B - Method, server, client and device for preventing message from being mistakenly sent - Google Patents
Method, server, client and device for preventing message from being mistakenly sent Download PDFInfo
- Publication number
- CN110391973B CN110391973B CN201910696016.3A CN201910696016A CN110391973B CN 110391973 B CN110391973 B CN 110391973B CN 201910696016 A CN201910696016 A CN 201910696016A CN 110391973 B CN110391973 B CN 110391973B
- Authority
- CN
- China
- Prior art keywords
- message
- conversation
- sent
- objects
- sending
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 76
- 230000015654 memory Effects 0.000 claims description 22
- 230000005540 biological transmission Effects 0.000 claims description 5
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
- H04L51/04—Real-time or near real-time messaging, e.g. instant messaging [IM]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
- H04L51/21—Monitoring or handling of messages
- H04L51/216—Handling conversation history, e.g. grouping of messages in sessions or threads
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
- H04L51/21—Monitoring or handling of messages
- H04L51/224—Monitoring or handling of messages providing notification on incoming messages, e.g. pushed notifications of received messages
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
- H04L51/52—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail for supporting social networking services
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computing Systems (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The invention discloses a method for preventing message from being mistakenly sent, which is applied to a server or a client and comprises the following steps: constructing a conversation topic model according to the historical information, wherein the conversation topic model comprises different conversation objects, conversation topics and corresponding relations between the conversation objects and the conversation topics; acquiring a message to be sent by a client, and analyzing the theme of the message to be sent; matching the topic of the message to be sent with the conversation topics of all the conversation objects in the conversation topic model, and sending corresponding prompt information to the client according to the matching result. Correspondingly, a server, a client and a device for preventing message missending are also provided. The method and the device are used for solving the problem that in the prior art, messages are easily sent to wrong receiving objects, and troubles are caused to senders.
Description
Technical Field
The present invention relates to the field of communications technologies, and in particular, to a method, a server, a client, and an apparatus for preventing message missending.
Background
When instant messages are sent through social software at present, sometimes the messages are sent to wrong receiving objects carelessly. For example, a message that should have been sent to A is sent inadvertently to B; or a message that should have been sent to group a is inadvertently sent to group B. The above situation mainly occurs due to the following reasons: the head portraits of the contacts are similar; the group names are similar; the method is characterized in that conversation is carried out with two (or more) recipients (groups) simultaneously in a short time, and misoperation is caused when a conversation interface is switched.
Although social software has a message revocation function, the content of a message may already be seen by the other party before the revocation operation is performed, and the revocation operation cannot be performed after the revocation time limit is exceeded. Therefore, such a situation that a message is easily sent to an incorrect recipient causes a certain trouble to the sender.
Disclosure of Invention
The present invention provides a method, a server, a client and a device for preventing message from being mistakenly sent, so as to solve the problem that the message is easily sent to the wrong receiving object in the prior art.
In order to solve the above technical problem, in a first aspect, the present invention provides a method for preventing a message from being mistakenly sent, where the method is applied to a server or a client, and the method includes:
s11, constructing a conversation topic model according to the historical information, wherein the conversation topic model comprises different conversation objects, conversation topics and corresponding relations between the conversation objects and the conversation topics;
s12, acquiring the message to be sent by the client, and analyzing the theme of the message to be sent;
and S13, matching the topic of the message to be sent with the conversation topics of all the conversation objects in the conversation topic model, and sending corresponding prompt information to the client according to the matching result.
Optionally, in step S13, sending corresponding prompt information to the client according to the matching result specifically includes:
s131, if the topic of the message to be sent is only matched with the conversation topic of the current conversation object or is not matched with the conversation topics of all the conversation objects, sending information of 'no need of prompting' to the client;
s132, if the topic of the message to be sent is matched with the conversation topics of a plurality of different conversation objects including the current conversation object, acquiring n conversation objects with the top matching sequence and corresponding matching degrees, and sending the n conversation objects to the client;
s133, if the conversation topic forming the matching relation with the topic of the message to be sent does not belong to the current conversation object, acquiring a second conversation object to which the message to be sent is expected to belong, and sending information of 'please prompt' and the name of the second conversation object to the client.
Optionally, step S11 specifically includes:
s111, listing all the messages related to the expression of the 'error message' in the historical messages as the subject of the 'error message', wherein the subject of the 'error message' comprises all the word descriptions related to the error message,
s112, dividing other history messages except the subject of the 'error message' into a plurality of subjects according to the conversation context of different conversation objects,
s113, constructing a conversation topic model according to the topics obtained in the step S111 and the step S112;
between step S12 and step S13, the method further comprises: judging whether the theme of the message to be sent is matched with the theme of the 'mistaken message': if yes, re-dividing the topics of m historical messages before the message to be sent, and correcting the conversation topic model in the step S113; if not, the flow proceeds to step S13,
after step S13, the method further includes:
when the operation of the user to withdraw the message is acquired, the withdrawn message is re-divided into topics, and the conversation topic model in the step S113 is modified;
when a message of "the prompt is wrong" is received, the message to be sent is re-divided into topics, and the dialogue topic model in step S113 is modified,
in step S12, analyzing the subject of the message to be sent includes:
s121, if the message is a link of an Internet webpage, taking webpage content as a subject of the message to be sent, and then entering the step S13;
s122, if the message is a file capable of analyzing the text content, taking the file content as the subject of the message to be sent, and then entering the step S13;
s123, if the message is a picture/video, executing the following steps:
s1231, extracting characters in the picture/video as a subject of the message to be sent, and then entering the step S13;
s1232, if no character can be extracted, or if no matching conversation topic and conversation object can be found in the topic of the message obtained in the step S1231 in the step S13, searching keywords related to the request to send pictures/videos in the conversation content of all the conversation objects in the latest time period, and proceeding to the step S14,
step S14 includes:
if the found keyword is only sent by the current conversation object or the keyword cannot be found, sending information of 'no need of prompting' to the client;
if a plurality of keywords of different conversation objects are found and are respectively sent by the different conversation objects including the current conversation object, sequencing the matching degrees of the keywords according to the sequence of the sending time of the keywords, wherein the sequencing of the matching degrees is closer to the front when the time is closer, and sending the sequenced conversation objects and the corresponding matching degrees thereof to the client;
and if the conversation object which is in the matching relation with the searched keyword does not belong to the current conversation object, acquiring one or more conversation objects to which the keyword belongs, and sending information of 'please prompt' and the name of one or more conversation objects to the client.
In a second aspect, the present invention provides a method for preventing a message from being mistakenly sent, which is applied to a client, and the method includes:
s21, receiving the message to be sent input by the user and sending the message to the server;
and S22, receiving the corresponding prompt information sent by the server, and outputting the prompt according to the prompt information.
Optionally, step S22 specifically includes:
when receiving the information of 'no need of prompting', not outputting the prompt;
when n conversation objects and corresponding matching degrees are received, the following steps are carried out:
s221, when it is obtained that the user has performed a session alternately between the current session object and any third session object of the received n session objects, and the matching degree with the third session object is higher than the matching degree with the current session object, outputting a prompt to "the matching degree of the message with the current session is lower than the matching degree with the third session object, please confirm whether the received object is selected incorrectly", and ask the user to confirm whether to "continue sending" or "reselect the sending object";
s222, when the matching degree difference of the received n conversation objects is small, outputting a prompt that the matching degree of the message and the current conversation object is ranked at the kth position, and asking to confirm whether the receiving object is selected wrongly, and displaying names of k-1 conversation objects ranked at the front, wherein k is more than 1 and less than or equal to n, and asking the user to confirm whether to continue to send or reselect the sending object;
when receiving the information of 'please prompt' and the name of one or more dialog objects, outputting 'the prompt of whether the message matches the current dialog with the degree lower than the degree of matching with the one or more dialog objects, please confirm whether the receiving object is selected wrongly' and ask the user to confirm whether 'continue sending' or 'reselect the sending object';
the method further comprises the following steps:
when the operation that the user selects 'continue sending' is acquired, the information that 'the prompt is wrong' is sent to the server.
In a third aspect, the present invention provides a method for preventing missending of messages, the method comprising a method applied to client application software and a method applied to a client human-machine exchange interface,
the method applied to the client application software adopts the method of the first aspect;
the method applied to the client man-machine exchange interface specifically comprises the following steps:
s31, receiving the message to be sent input by the user and sending the message to the client application software;
and S32, receiving corresponding prompt information sent by the client application software, and outputting a prompt according to the prompt information.
Optionally, step S32 specifically includes:
when receiving the information of 'no need of prompting', not outputting the prompt;
when n conversation objects and corresponding matching degrees are received, the following steps are carried out:
s221, when it is obtained that the user has performed a session alternately between the current session object and any third session object of the received n session objects, and the matching degree with the third session object is higher than the matching degree with the current session object, outputting a prompt to "the matching degree of the message with the current session is lower than the matching degree with the third session object, please confirm whether the received object is selected incorrectly", and ask the user to confirm whether to "continue sending" or "reselect the sending object";
s222, when the matching degree difference of the received n conversation objects is small, outputting a prompt that the matching degree of the message and the current conversation object is ranked at the kth position, and asking to confirm whether the receiving object is selected wrongly, and displaying names of k-1 conversation objects ranked at the front, wherein k is more than 1 and less than or equal to n, and asking the user to confirm whether to continue to send or reselect the sending object;
when receiving the information of 'please prompt' and the name of one or more dialog objects, outputting 'the prompt of whether the message matches with the current dialog with a lower degree than the one or more dialog objects, please confirm whether the receiving object is selected wrongly' and ask the user to confirm whether to 'continue sending' or 'reselect the sending object',
the method further comprises the following steps:
when receiving the operation of selecting 'continue sending' by the user, sending 'the prompt is wrong' information to the client application software;
when receiving the operation of selecting 'reselecting a transmission object' by a user, executing the following steps:
s223, copying the message to be sent to a system pasting board;
s224, after the user reselects the sending object, popping up a paste button in the message input box;
and S225, when the user clicks the paste button, pasting the message copied into the system paste board into the message input box.
In a fourth aspect, the present invention provides a server comprising a memory and a processor, the memory storing instructions that the processor executes to perform the method as applied to the server in the first aspect.
In a fifth aspect, the present invention provides a client comprising a memory and a processor, the memory storing instructions that the processor executes to perform the method of the second aspect as applied to the client.
In a sixth aspect, the present invention provides an apparatus for preventing missent messages, the apparatus comprising a message analysis module and a message alert module,
the message analysis module is connected with the message reminding module and used for executing the following steps:
constructing a conversation topic model according to the historical information, wherein the conversation topic model comprises different conversation objects, conversation topics and corresponding relations between the conversation objects and the conversation topics;
receiving a message to be sent by a message reminding module, and analyzing the theme of the message to be sent;
matching the topic of the message to be sent with the conversation topics of all conversation objects in the conversation topic model, and sending corresponding prompt information to a message reminding module according to the matching result;
the message reminding module is used for executing the following steps:
receiving the message to be sent input by a user and sending the message to a message analysis module;
and receiving corresponding prompt information sent by the message analysis module, and outputting a prompt according to the prompt information.
The beneficial technical effects of the invention are as follows:
according to the technical scheme of the method, the server, the client and the device for preventing the message from being mistakenly sent, the topic analysis is firstly carried out on the history record of the instant message according to the conversation context of different conversation objects (contacts or groups), and a conversation topic model is constructed. The subsequent messages are subject analyzed before being sent to the addressee. Matching the topic of the message to be sent with the conversation topics of all the conversation objects in the conversation topic model, and sending corresponding prompt information to the client according to the matching result:
if the theme of the message to be sent is only matched with the conversation theme of the current conversation object or is not matched with the conversation themes of all the conversation objects, sending information of 'no need of prompting' to the client, and the client does not output prompting information;
if the topic of the message to be sent is matched with the conversation topics of a plurality of different conversation objects including the current conversation object, acquiring n conversation objects which are matched and ranked at the top and corresponding matching degrees, and sending the matching degrees to the client, wherein the client has prompt contents under two conditions when receiving the n conversation objects and corresponding matching degree information:
firstly, when it is acquired that the user has performed a conversation alternately between the current conversation object and any third conversation object in the received n conversation objects and the matching degree with the third conversation object is higher than that with the current conversation object, a prompt of "the matching degree of the message with the current conversation is lower than that with the third conversation object, and whether the received object is selected by mistake is requested to be confirmed, and the user is requested to confirm whether to" continue sending "or" reselect the sending object "is output;
secondly, when the matching degree difference of the received n conversation objects is smaller, outputting a prompt of arranging the matching degree of the message and the current conversation object at the kth position and asking for confirming whether the receiving object is selected wrongly, and displaying names of k-1 conversation objects which are ranked at the front, wherein 1< k is less than or equal to n, and asking the user for confirming whether to continue to send or reselect the sending object,
if the conversation theme forming a matching relationship with the theme of the message to be sent does not belong to the current conversation object, acquiring one or more conversation objects to which the message to be sent is expected to belong, sending information of ' please prompt ' and the names of the one or more conversation objects to the client, and the client outputs a prompt of ' the matching degree of the message and the current conversation is lower than the matching degree of the message and the one or more conversation objects, please confirm whether to select the receiving object wrongly ' and ask the user to confirm whether to continue sending or reselect the sending object '.
According to the technical scheme, the message to be sent is subjected to theme analysis and matched with the theme of the historical message, the relevant prompt information is output according to the matching result, and the user is prompted before the user sends the message, so that the user can be prevented from sending the message to a wrong receiving object, and the trouble on the sender is avoided.
Drawings
FIG. 1: a method for preventing message missending in embodiment 1 of the present invention is a flowchart (applied to a server or a client);
FIG. 2: a method for preventing message missending in embodiment 2 of the present invention is a flowchart (applied to a client);
FIG. 3: a schematic structural diagram of a device for preventing missending of a message according to embodiment 6 of the present invention;
FIG. 4: the invention provides a prompt message display interface of a message reminding module in embodiment 6.
Detailed Description
In order to make those skilled in the art better understand the technical solution of the present invention, the method, the server, the client and the device for preventing the message from being mistakenly sent of the present invention are described in further detail below with reference to the accompanying drawings and embodiments.
Example 1:
as shown in fig. 1, the present embodiment provides a method for preventing a message from being mistakenly sent, which is applied to a server or a client, and the method includes:
s11, constructing a conversation topic model according to the historical information, wherein the conversation topic model comprises different conversation objects, conversation topics and corresponding relations between the conversation objects and the conversation topics;
the method for constructing the conversation topic model specifically comprises the following steps:
and S111, individually listing all the messages related to the expression of the 'mistaken messages' in the history messages as the topic of the 'mistaken messages', wherein the topic of the 'mistaken messages' comprises all the text descriptions related to the mistaken messages, such as 'mistaken sending' and the like.
S112, dividing the history messages except the "missed message" topic into a plurality of topics according to the conversation context of different conversation objects (contacts or groups), for example, a common clustering algorithm may be adopted, which includes a division method, a hierarchy method, a density-based method, and the like, and in this embodiment, a clustering algorithm is preferably adopted.
Since the clustering algorithm is well known to those skilled in the art, the detailed partitioning process will not be described in detail here.
And S113, constructing a conversation topic model according to the topics obtained in the step S111 and the step S112.
In this embodiment, the "missed message" topic may be obtained according to step S111, other topics except the "missed message" topic may be obtained according to step S112, and then the conversation topic model is constructed according to the "missed message" topic and the other topics. In the constructed conversation topic model, for a separately listed "missed message" topic, which includes all the word descriptions about the missed messages in the history messages, it should be noted that the topic has no corresponding conversation object. In addition, the creation of the conversation topic model takes the form of dynamic updates. Namely, when each conversation object has new information, the new information is used as the input of the established conversation topic model, the selected clustering algorithm is executed again, and the model parameters of the established conversation topic model are updated, so that the dynamic correction of the established conversation topic model is completed.
S12, acquiring the message to be sent by the client, and analyzing the theme of the message to be sent;
optionally, between step S12 and step S13, the method further comprises: judging whether the theme of the message to be sent is matched with the theme of the 'mistaken message': if yes, re-dividing the topic of m history messages before the message to be sent (where m is a settable parameter, for example, m may be set to 3, which indicates that 3 history messages before the message to be sent are present), i.e. re-performing the clustering algorithm on the m history messages, and modifying the conversation topic model in step S113; if not, the flow proceeds to step S13,
and S13, matching the topic of the message to be sent with the conversation topics of all the conversation objects in the conversation topic model, and sending corresponding prompt information to the client according to the matching result.
In this embodiment, sending the corresponding prompt information to the client according to the matching result specifically includes:
s131, if the topic of the message to be sent is only matched with the conversation topic of the current conversation object or is not matched with the conversation topics of all the conversation objects, sending information of 'no need of prompting' to the client;
s132, if the topic of the message to be sent matches the conversation topics of multiple different conversation objects including the current conversation object, acquiring n (where n is a settable parameter, for example, n may be set to 3, which indicates that the first 3 bits with the highest matching degree are acquired) conversation objects ranked at the top of the matching sequence and corresponding matching degrees (where the matching degrees are expressed by percentages), and sending the n conversation objects and the corresponding matching degrees to the client;
s133, if all conversation themes forming a matching relation with the theme of the message to be sent do not belong to the current conversation object, acquiring a second conversation object to which the message to be sent is expected to belong, and sending information of 'please prompt' and the name of the second conversation object to the client.
Optionally, after step S13, the method for preventing message missending further includes:
when an operation of retrieving the message withdrawal from the user is obtained, for example, the user uses a "withdrawal" command, the topic is subdivided for the withdrawn message, in this embodiment, the clustering operation is performed again for the withdrawn message, and the conversation topic model in step S113 is modified;
when a message "this prompt is wrong" is received (indicating that the user has selected to continue sending the message), the message to be sent in step S12 is re-divided into topics, in this embodiment, the clustering operation is performed on the message again, and the conversation topic model in step S113 is modified.
Optionally, in step S12, analyzing the subject of the message to be sent includes:
s121, if the message is a link of an Internet webpage, taking webpage content as a subject of the message to be sent, and then entering the step S13;
s122, if the message is a file (e.g. word, excel, pdf, etc.) capable of parsing text content, taking the file content as the subject of the message to be sent, and then entering step S13;
s123, if the message is a picture/video, executing the following steps:
s1231, extracting characters in the picture/video as a subject of the message to be sent, and then entering the step S13;
in this embodiment, the picture is preferably a screenshot of a social software chat record or a picture with other text information as a main body, so that the text in the picture can be extracted as a subject of the message to be sent. The video is preferably a short video shot by social software, and therefore, characters in the video cannot be extracted in some cases.
S1232, if no text can be extracted, or no matching conversation topic and conversation object can be found in the topic of the message obtained in step S1231 in step S13, finding out keywords related to the picture/video requested to be sent (for example, the keywords related to the picture requested to be sent may be "screenshot", "photo", "picture", etc., and the keywords related to the video requested to be sent may be "video", "short" etc.) in the conversation content in the latest time period (the time period is a settable parameter, for example, the time period may be set to 10 minutes) of all conversation objects, and proceeding to step S14,
step S14 includes:
if the found keyword is only sent by the current conversation object or the keyword cannot be found, sending information of 'no need of prompting' to the client;
if a plurality of keywords of different conversation objects are found and are respectively sent by the different conversation objects including the current conversation object, the keywords are sorted according to the sequence of the sending time of the keywords, the more the time is, the more the matching degree is, the more the sequence is, and the sorted conversation objects and the corresponding matching degrees are sent to the client.
For example, if the keyword "screenshot" is found in the last 4 minutes of the dialog content with the current dialog object a, the keyword "screenshot" is found in the last 3 minutes of the dialog content with the dialog object B, and the keyword "picture" is found in the last 1 minute of the dialog content with the dialog object C, the matching degree of the dialog object C is the highest, the matching degree of the dialog object B is the second, and the matching degree of the current dialog object a is the lowest, the dialog object C and the matching degree thereof, the dialog object B and the matching degree thereof, and the dialog object a and the matching degree thereof are sent to the client.
And if the conversation object which is in the matching relation with the searched keyword does not belong to the current conversation object, acquiring one or more conversation objects to which the keyword belongs, and sending information of 'please prompt' and the name of one or more conversation objects to the client.
For example, if the current dialog object is a and the dialog object in matching relationship with the found keyword is B, sending information of 'please prompt' and the name of the dialog object B expected to belong to the client; if the current conversation object is A, a keyword screenshot is found in the conversation content within the last 3 minutes of the conversation object B, a keyword picture is found in the conversation content within the last 1 minute of the conversation object C, and two forms of prompt information can be sent to the client.
Example 2:
as shown in fig. 2, the present embodiment provides a method for preventing a message from being mistakenly sent, which is applied to a client, and the method includes:
s21, receiving the message to be sent input by the user and sending the message to the server;
and S22, receiving the corresponding prompt information sent by the server, and outputting the prompt according to the prompt information.
Optionally, step S22 specifically includes:
when receiving the information of 'no need of prompting' sent by the server, the client does not output the prompt;
when n (same as the parameter n in embodiment 1) session objects and corresponding matching degrees sent by the server are received, the following steps are performed:
s221, when it is obtained that the user has performed a session alternately between the current session object and any third session object of the received n session objects, and the matching degree with the third session object is higher than the matching degree with the current session object, outputting a prompt to "the matching degree of the message with the current session is lower than the matching degree with the third session object, please confirm whether the received object is selected incorrectly", and ask the user to confirm whether to "continue sending" or "reselect the sending object";
for example, it is obtained that the user has alternated between the current session object a (e.g. a contact or a group) and any one session object B of the received n session objects, and the matching degree with the session object B is higher than that with the current session object a, a prompt of "the matching degree of the message with the current session is lower than that with the session object B, please confirm whether the receiving object is selected mistakenly" is output, and the user is asked to confirm whether to "continue sending" or "reselect the sending object".
S222, when the matching degree difference between n received dialog objects is small (for example, the matching degree difference between any two of the two matching objects is smaller than a threshold, and the threshold may be set to 5% or other percentage values), outputting a prompt of "the matching degree of the message with the current dialog object is ranked at the kth position, and please confirm whether the received object is selected incorrectly", and displaying names of k-1 dialog objects ranked at the top, where 1< k ≦ n, and asking the user to confirm whether to "continue sending" or "reselect the sending object";
when receiving the information of 'please prompt' and the name of one or more dialog objects, outputting 'the prompt of whether the message matches the current dialog with the degree lower than the degree of matching with the one or more dialog objects, please confirm whether the receiving object is selected wrongly' and ask the user to confirm whether 'continue sending' or 'reselect the sending object';
in this embodiment, a preferable mode is further provided, and when the client determines that the most likely message receiving object is X according to the received information, a prompt of "please confirm whether you actually want to send the message to X" is output, and the user is asked to confirm whether "continue sending" or "send the message to X". When the user clicks "continue sending" the message will be sent to the current dialog object (indicating that the user decides there is no error in sending the object); when the user clicks "send message to X", the client sends the message directly to X.
Optionally, the method further comprises:
when the operation that the user selects 'continue sending' is acquired, the information that 'the prompt is wrong' is sent to the server. When the client side obtains the operation of selecting 'continue sending' by the user and indicates that the adopted conversation topic model analysis is wrong, the client side sends the 'prompt is wrong' information to the server, so that the server can perform clustering operation on the corresponding information again and correct the established conversation topic model.
Example 3:
the embodiment provides a method for preventing message missending, which comprises a method applied to client application software and a method applied to a client man-machine exchange interface,
the method applied to the client application software adopts the method applied to the client in the embodiment 1, wherein the client application software can be social contact software or instant messaging software;
the method applied to the client man-machine exchange interface specifically comprises the following steps:
s31, receiving the message to be sent input by the user and sending the message to the client application software;
in this embodiment, a user edits a message on a client-side human-machine exchange interface and clicks the message to send, the client-side human-machine exchange interface receives the message and sends the message to client-side application software, and the client-side application software performs real-time theme analysis on the message by using the method described in embodiment 1 and sends corresponding prompt information to the client-side human-machine exchange interface according to the analysis result.
And S32, receiving corresponding prompt information sent by the client application software, and outputting a prompt according to the prompt information.
Optionally, step S32 specifically includes:
when receiving the information of 'no need of prompting', not outputting the prompt;
when n conversation objects and corresponding matching degrees are received, the following steps are carried out:
s221, when it is obtained that the user has performed a session alternately between the current session object and any third session object of the received n session objects, and the matching degree with the third session object is higher than the matching degree with the current session object, outputting a prompt to "the matching degree of the message with the current session is lower than the matching degree with the third session object, please confirm whether the received object is selected incorrectly", and ask the user to confirm whether to "continue sending" or "reselect the sending object";
s222, when the matching degree difference of the received n conversation objects is small, outputting a prompt that the matching degree of the message and the current conversation object is ranked at the kth position, and asking to confirm whether the receiving object is selected wrongly, and displaying names of k-1 conversation objects ranked at the front, wherein k is more than 1 and less than or equal to n, and asking the user to confirm whether to continue to send or reselect the sending object;
when receiving the information of 'please prompt' and the name of one or more dialog objects, outputting 'the prompt of whether the message matches with the current dialog with a lower degree than the one or more dialog objects, please confirm whether the receiving object is selected wrongly' and ask the user to confirm whether to 'continue sending' or 'reselect the sending object',
optionally, the method further comprises:
when receiving the operation of selecting 'continue sending' (indicating that the topic clustering analysis is wrong), sending 'the prompt is wrong' information to the client application software so that the client application software corrects the conversation topic model;
when receiving the operation of selecting 'reselecting a transmission object' by a user, executing the following steps:
s223, copying the message to be sent to a system pasting board;
s224, after the user reselects the sending object, popping up a paste button in the message input box;
s225, when the user clicks the paste button, the message copied to the system paste board is pasted to the message input box, and the user clicks the sending button to finish message sending.
Example 4:
the present embodiment provides a server, including a memory and a processor, the memory storing instructions, and the processor executing the instructions to execute the method for preventing missent messages applied to the server in embodiment 1.
The memory is electrically connected with the processor, the memory can adopt a flash memory or a read-only memory or other memories, and the processor can adopt a central processing unit or a singlechip.
Example 5:
the present embodiment provides a client, including a memory and a processor, where the memory stores instructions, and the processor executes the instructions to execute the method for preventing message missending provided in embodiment 2.
The memory is electrically connected with the processor, the memory can adopt a flash memory or a read-only memory or other memories, and the processor can adopt a central processing unit or a singlechip.
Example 6:
as shown in fig. 3, the present embodiment provides a device for preventing missent messages, which may be integrated inside client-side social software or instant messaging software, the device includes a message analysis module 1 and a message reminding module 2,
the message analysis module 1 is connected with the message reminding module 2, and the message analysis module 1 is used for executing the following steps:
constructing a conversation topic model according to the historical information, wherein the conversation topic model comprises different conversation objects, conversation topics and corresponding relations between the conversation objects and the conversation topics;
receiving a message to be sent by the message reminding module 2, and analyzing the theme of the message to be sent;
matching the topic of the message to be sent with the conversation topics of all conversation objects in the conversation topic model, and sending corresponding prompt information to the message reminding module 2 according to the matching result;
in this embodiment, the detailed method for constructing the conversation topic model by the message analysis module 1 is as described in embodiment 1, and is not described herein again. The message analysis module 1 is first configured to determine whether a topic of the message to be sent matches a "missed message" topic: if yes, re-dividing the topic of m historical messages before the message to be sent (where m is a settable parameter, for example, m may be set to 3, which indicates that 3 historical messages before the message to be sent are present), i.e. re-performing the clustering algorithm on the m historical messages, and modifying the conversation topic model; if not, matching the topic of the message to be sent with the conversation topics of all the conversation objects in the conversation topic model, and sending corresponding prompt information to the message reminding module 2 according to the matching result.
The prompt message specifically includes:
if the topic of the message to be sent is only matched with the conversation topic of the current conversation object or is not matched with the conversation topics of all the conversation objects, sending information of 'no need of prompting' to the message reminding module 2;
if the topic of the message to be sent matches the conversation topics of a plurality of different conversation objects including the current conversation object at the same time, acquiring n (where n is a settable parameter, for example, n may be set to 3, which indicates that the top 3 bits with the highest matching degree are acquired) conversation objects with the top ranking and corresponding matching degrees (where the matching degrees are expressed in percentage), and sending the n conversation objects and the corresponding matching degrees to the message reminding module 2;
if all conversation themes forming a matching relation with the theme of the message to be sent do not belong to the current conversation object, acquiring a second conversation object to which the message to be sent is expected to belong, and sending information of 'please prompt' and the name of the second conversation object to the message reminding module 2.
The message reminding module 2 is used for executing the following steps:
receiving a message to be sent input by a user and sending the message to the message analysis module 1;
and receiving corresponding prompt information sent by the message analysis module 1, and outputting a prompt according to the prompt information. The method comprises the following specific steps:
when receiving the information of 'no need of prompting' sent by the message analysis module 1, the message reminding module 2 does not output the prompt;
when receiving n dialog objects of the message analysis module 1 and the corresponding matching degrees, the following steps are performed:
when the fact that the user has alternately performed conversation between the current conversation object and any third conversation object in the received n conversation objects and the matching degree of the user and the third conversation object is higher than the matching degree of the user and the current conversation object is obtained, a prompt of 'whether the matching degree of the message and the current conversation is lower than the matching degree of the message and the third conversation object and whether the receiving object is selected mistakenly is confirmed, and the user is asked to confirm whether to continue to send or reselect the sending object' is output;
when the matching degree difference of the received n conversation objects is small (for example, the matching degree difference between any two matching objects is smaller than a threshold value, and the threshold value can be set to be 5% or other percentage values), a prompt that "the matching degree of the message and the current conversation object is arranged at the kth position, and whether a receiving object is selected wrongly or not" is output, and names of k-1 conversation objects which are ranked at the top are displayed, wherein 1< k is less than or equal to n, and a user is requested to confirm whether to "continue to send" or "reselect a sending object";
when receiving the information of 'please prompt' and the name of one or more dialog objects, outputting 'the prompt of whether the message matches with the current dialog with a lower degree than the one or more dialog objects, please confirm whether the receiving object is selected wrongly' and ask the user to confirm whether 'continue to send' or 'reselect the sending object'.
In this embodiment, the interface of the message reminding module 2 for outputting the prompt message is shown in fig. 4, and includes three information boxes, where the information box 3 displays specific prompt contents, where the specific prompt contents are "the matching degree of the message with the current dialog is lower than the matching degree with the third dialog object, and please confirm whether the receiving object is selected incorrectly", or "the matching degree of the message with the current dialog object is ranked at the kth position, and please confirm whether the receiving object is selected incorrectly", or "the matching degree of the message with the current dialog is lower than the matching degree with one or more dialog objects, and please confirm whether the receiving object is selected incorrectly". The information frame 4 displays "continue transmission", and the information frame 5 displays "reselect transmission target", wherein the information frame 4 and the information frame 5 are buttons and can accept input operation of a user.
The beneficial effects of examples 1-6 are:
according to the technical scheme of the method, the server, the client and the device for preventing the message from being mistakenly sent, the topic analysis is firstly carried out on the history of the instant message according to the conversation context of different conversation objects (contacts or groups), and a conversation topic model is constructed. Before the message is sent to the receiver, the message to be sent is subject analyzed, the subject of the message to be sent is matched with the conversation subjects of all the conversation objects in the conversation subject model, and corresponding prompt information is output to a user according to the matching result, namely, the sender is prompted for possible errors before the message is sent out, and the message sending operation is executed after the sender confirms the message. After the message is successfully sent to the addressee, the message becomes a new history record of the instant message, a new history message is formed, and then the conversation topic model is continuously corrected, so that the conversation topic model is more perfect and stable, and the prompt message which is subjected to matching analysis through the conversation topic model is more accurate, so that the message can be effectively prevented from being sent to a wrong receiving object, and the sending person is prevented from causing troubles.
It will be understood that the above embodiments are merely exemplary embodiments taken to illustrate the principles of the present invention, which is not limited thereto. It will be apparent to those skilled in the art that various modifications and improvements can be made without departing from the spirit and substance of the invention, and these modifications and improvements are also considered to be within the scope of the invention.
Claims (10)
1. A method for preventing message from being sent by mistake, which is applied to a server or a client, and is characterized in that the method comprises the following steps:
s11, constructing a conversation topic model according to the historical information, wherein the conversation topic model comprises different conversation objects, conversation topics and corresponding relations between the conversation objects and the conversation topics;
s12, acquiring the message to be sent by the client, and analyzing the theme of the message to be sent;
s13, matching the topic of the message to be sent with the conversation topics of all the conversation objects in the conversation topic model, sending corresponding prompt information to the client according to the matching result,
step S11 specifically includes:
s111, listing all the messages related to the expression of the 'error message' in the historical messages as the subject of the 'error message', wherein the subject of the 'error message' comprises all the word descriptions related to the error message,
s112, dividing other history messages except the subject of the 'error message' into a plurality of subjects according to the conversation context of different conversation objects,
and S113, constructing a conversation topic model according to the topics obtained in the step S111 and the step S112.
2. The method for preventing the message from being mistakenly sent according to claim 1, wherein in step S13, sending the corresponding prompt message to the client according to the matching result specifically includes:
s131, if the topic of the message to be sent is only matched with the conversation topic of the current conversation object or is not matched with the conversation topics of all the conversation objects, sending information of 'no need of prompting' to the client;
s132, if the topic of the message to be sent is matched with the conversation topics of a plurality of different conversation objects including the current conversation object, acquiring n conversation objects with the top matching sequence and corresponding matching degrees, and sending the n conversation objects to the client;
s133, if the conversation topic forming the matching relation with the topic of the message to be sent does not belong to the current conversation object, acquiring a second conversation object to which the message to be sent is expected to belong, and sending information of 'please prompt' and the name of the second conversation object to the client.
3. The method of preventing misappropriated messages according to claim 2,
between step S12 and step S13, the method further comprises: judging whether the theme of the message to be sent is matched with the theme of the 'mistaken message': if yes, re-dividing the topics of m historical messages before the message to be sent, and correcting the conversation topic model in the step S113; if not, the flow proceeds to step S13,
after step S13, the method further includes:
when the operation of the user to withdraw the message is acquired, the withdrawn message is re-divided into topics, and the conversation topic model in the step S113 is modified;
when a message of "the prompt is wrong" is received, the message to be sent is re-divided into topics, and the dialogue topic model in step S113 is modified,
in step S12, analyzing the subject of the message to be sent includes:
s121, if the message is a link of an Internet webpage, taking webpage content as a subject of the message to be sent, and then entering the step S13;
s122, if the message is a file capable of analyzing the text content, taking the file content as the subject of the message to be sent, and then entering the step S13;
s123, if the message is a picture/video, executing the following steps:
s1231, extracting characters in the picture/video as a subject of the message to be sent, and then entering the step S13;
s1232, if no character can be extracted, or if no matching conversation topic and conversation object can be found in the topic of the message obtained in the step S1231 in the step S13, searching keywords related to the request to send pictures/videos in the conversation content of all the conversation objects in the latest time period, and proceeding to the step S14,
step S14 includes:
if the found keyword is only sent by the current conversation object or the keyword cannot be found, sending information of 'no need of prompting' to the client;
if a plurality of keywords of different conversation objects are found and are respectively sent by the different conversation objects including the current conversation object, sequencing the matching degrees of the keywords according to the sequence of the sending time of the keywords, wherein the sequencing of the matching degrees is closer to the front when the time is closer, and sending the sequenced conversation objects and the corresponding matching degrees thereof to the client;
and if the conversation object which is in the matching relation with the searched keyword does not belong to the current conversation object, acquiring one or more conversation objects to which the keyword belongs, and sending information of 'please prompt' and the name of one or more conversation objects to the client.
4. A method for preventing message from being sent by mistake is applied to a client side, and is characterized in that the method comprises the following steps:
s21, receiving a message to be sent input by a user, sending the message to a server, so that the server constructs a conversation topic model according to historical messages, wherein the conversation topic model comprises different conversation objects, conversation topics and corresponding relations between the conversation objects and the conversation topics, analyzing the topic of the message to be sent, matching the topic of the message to be sent with the conversation topics of all the conversation objects in the conversation topic model, and sending corresponding prompt information to a client according to the matching result;
the method for constructing the conversation topic model specifically comprises the following steps:
s111, listing all the messages related to the expression of the 'error message' in the historical messages as the subject of the 'error message', wherein the subject of the 'error message' comprises all the word descriptions related to the error message,
s112, dividing other history messages except the subject of the 'error message' into a plurality of subjects according to the conversation context of different conversation objects,
s113, constructing a conversation topic model according to the topics obtained in the step S111 and the step S112;
and S22, receiving the corresponding prompt information sent by the server, and outputting the prompt according to the prompt information.
5. The method according to claim 4, wherein the step S22 specifically comprises:
when receiving the information of 'no need of prompting', not outputting the prompt;
when n conversation objects and corresponding matching degrees are received, the following steps are carried out:
s221, when it is obtained that the user has performed a session alternately between the current session object and any third session object of the received n session objects, and the matching degree with the third session object is higher than the matching degree with the current session object, outputting a prompt to "the matching degree of the message with the current session is lower than the matching degree with the third session object, please confirm whether the received object is selected incorrectly", and ask the user to confirm whether to "continue sending" or "reselect the sending object";
s222, when the matching degree difference of the received n conversation objects is small, outputting a prompt that the matching degree of the message and the current conversation object is ranked at the kth position, and asking to confirm whether the receiving object is selected wrongly, and displaying names of k-1 conversation objects ranked at the front, wherein k is more than 1 and less than or equal to n, and asking the user to confirm whether to continue to send or reselect the sending object;
when receiving the information of 'please prompt' and the name of one or more dialog objects, outputting 'the prompt of whether the message matches the current dialog with the degree lower than the degree of matching with the one or more dialog objects, please confirm whether the receiving object is selected wrongly' and ask the user to confirm whether 'continue sending' or 'reselect the sending object';
the method further comprises the following steps:
when the operation that the user selects 'continue sending' is acquired, the information that 'the prompt is wrong' is sent to the server.
6. A method for preventing missending of messages, characterized in that the method comprises a method applied to client application software and a method applied to a client human-machine exchange interface,
the method applied to the client application software adopts the method of any one of claims 1-3;
the method applied to the client man-machine exchange interface specifically comprises the following steps:
s31, receiving the message to be sent input by the user and sending the message to the client application software;
and S32, receiving corresponding prompt information sent by the client application software, and outputting a prompt according to the prompt information.
7. The method according to claim 6, wherein step S32 specifically comprises:
when receiving the information of 'no need of prompting', not outputting the prompt;
when n conversation objects and corresponding matching degrees are received, the following steps are carried out:
s221, when it is obtained that the user has performed a session alternately between the current session object and any third session object of the received n session objects, and the matching degree with the third session object is higher than the matching degree with the current session object, outputting a prompt to "the matching degree of the message with the current session is lower than the matching degree with the third session object, please confirm whether the received object is selected incorrectly", and ask the user to confirm whether to "continue sending" or "reselect the sending object";
s222, when the matching degree difference of the received n conversation objects is small, outputting a prompt that the matching degree of the message and the current conversation object is ranked at the kth position, and asking to confirm whether the receiving object is selected wrongly, and displaying names of k-1 conversation objects ranked at the front, wherein k is more than 1 and less than or equal to n, and asking the user to confirm whether to continue to send or reselect the sending object;
when receiving the information of 'please prompt' and the name of one or more dialog objects, outputting 'the prompt of whether the message matches with the current dialog with a lower degree than the one or more dialog objects, please confirm whether the receiving object is selected wrongly' and ask the user to confirm whether to 'continue sending' or 'reselect the sending object',
the method further comprises the following steps:
when receiving the operation of selecting 'continue sending' by the user, sending 'the prompt is wrong' information to the client application software;
when receiving the operation of selecting 'reselecting a transmission object' by a user, executing the following steps:
s223, copying the message to be sent to a system pasting board;
s224, after the user reselects the sending object, popping up a paste button in the message input box;
and S225, when the user clicks the paste button, pasting the message copied into the system paste board into the message input box.
8. A server comprising a memory and a processor, the memory storing instructions that the processor executes to perform the method according to any one of claims 1-3.
9. A client comprising a memory and a processor, the memory storing instructions that the processor executes to perform the method according to any one of claims 4 to 5.
10. An apparatus for preventing missent messages, which comprises a message analysis module and a message reminding module,
the message analysis module is connected with the message reminding module and used for executing the following steps:
constructing a conversation topic model according to the historical information, wherein the conversation topic model comprises different conversation objects, conversation topics and corresponding relations between the conversation objects and the conversation topics;
receiving a message to be sent by a message reminding module, and analyzing the theme of the message to be sent;
matching the topic of the message to be sent with the conversation topics of all the conversation objects in the conversation topic model, and sending corresponding prompt information to a message reminding module according to the matching result, wherein the message reminding module is used for executing the following steps:
receiving the message to be sent input by a user and sending the message to a message analysis module;
receiving corresponding prompt information sent by the message analysis module, outputting a prompt according to the prompt information,
specifically, the message analysis module is configured to list, as a topic of the "missed message", all messages in the history messages that relate to the expression of the "missed message", wherein the topic of the "missed message" includes all text descriptions related to the missed message, divide the history messages excluding the topic of the "missed message" into a plurality of topics according to the conversation contexts of different conversation objects, and construct a conversation topic model according to the obtained topics.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910696016.3A CN110391973B (en) | 2019-07-30 | 2019-07-30 | Method, server, client and device for preventing message from being mistakenly sent |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910696016.3A CN110391973B (en) | 2019-07-30 | 2019-07-30 | Method, server, client and device for preventing message from being mistakenly sent |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110391973A CN110391973A (en) | 2019-10-29 |
CN110391973B true CN110391973B (en) | 2021-09-10 |
Family
ID=68288079
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910696016.3A Active CN110391973B (en) | 2019-07-30 | 2019-07-30 | Method, server, client and device for preventing message from being mistakenly sent |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110391973B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111158838B (en) * | 2019-12-31 | 2022-05-31 | 联想(北京)有限公司 | Information processing method and device |
CN112540712B (en) * | 2020-12-07 | 2022-05-10 | 维沃移动通信有限公司 | Message processing method and device, electronic equipment and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105099853A (en) * | 2014-04-25 | 2015-11-25 | 国际商业机器公司 | Erroneous message sending preventing method and system |
CN106375377A (en) * | 2016-08-25 | 2017-02-01 | 深圳市金立通信设备有限公司 | Data processing method and terminal |
CN108566330A (en) * | 2018-03-21 | 2018-09-21 | 联想(北京)有限公司 | Information processing method and the first electronic equipment |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20190019227A1 (en) * | 2017-07-11 | 2019-01-17 | Harold T. Fogg | System and method for streamlining access to device information |
-
2019
- 2019-07-30 CN CN201910696016.3A patent/CN110391973B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105099853A (en) * | 2014-04-25 | 2015-11-25 | 国际商业机器公司 | Erroneous message sending preventing method and system |
CN106375377A (en) * | 2016-08-25 | 2017-02-01 | 深圳市金立通信设备有限公司 | Data processing method and terminal |
CN108566330A (en) * | 2018-03-21 | 2018-09-21 | 联想(北京)有限公司 | Information processing method and the first electronic equipment |
Also Published As
Publication number | Publication date |
---|---|
CN110391973A (en) | 2019-10-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10460029B2 (en) | Reply information recommendation method and apparatus | |
CN109308357B (en) | Method, device and equipment for obtaining answer information | |
CN112540712B (en) | Message processing method and device, electronic equipment and storage medium | |
CN104795069B (en) | Speech recognition method and server | |
CN110391973B (en) | Method, server, client and device for preventing message from being mistakenly sent | |
US20080320086A1 (en) | System and method for updating instant message transcripts | |
CN111147268B (en) | Group chatting method and device, electronic equipment and storage medium | |
US20230396567A1 (en) | Chat bot utilizing metaphors to both relay and obtain information | |
CN105323145B (en) | Malicious information identification method, malicious information identification device and system | |
CN105763535B (en) | Conversation message list processing method, equipment and system | |
WO2013112958A1 (en) | Processing data relating to a communication event | |
CN106911562B (en) | A kind of information correction reminding method and its device in instant messaging | |
CN111010335A (en) | Chat expression sending method and device, electronic equipment and medium | |
CN109858923B (en) | Man-machine conversation method and device based on robot | |
CN113836287A (en) | Method and device for determining question and answer result, computer equipment and storage medium | |
CN111158838B (en) | Information processing method and device | |
CN111400475A (en) | Information processing method and device and electronic equipment | |
CN113032538A (en) | Topic transfer method based on knowledge graph, controller and storage medium | |
CN113271247B (en) | Information processing method, apparatus, device and storage medium | |
CN117010725B (en) | Personalized decision method, system and related device | |
CN113438280B (en) | Vehicle starting control method and device | |
CN114579712B (en) | Text attribute extraction and matching method based on dynamic model | |
JP4688629B2 (en) | Electronic message classification device and electronic message classification method | |
CN113963687A (en) | Voice interaction method, device, equipment and storage medium | |
CN106502488B (en) | Interaction method and device based on content input |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |