CN108900687A - It breaks one's promise the display methods and device of number - Google Patents

Abstract

The invention discloses the display methods and device of a kind of number of breaking one's promise.Wherein, this method includes：Word segmentation processing is carried out to the target message got, feature vector corresponding with the target message is determined according to word segmentation processing result；Described eigenvector is inputted into preset SMS classified model, when the output result of the SMS classified model be break one's promise type result when, communicating number corresponding with the target message is screened as destination number；Judge whether the destination number is number of breaking one's promise, if so, for the break one's promise label of the number setting for being shown in communication of breaking one's promise.Which label that can will break one's promise when object of breaking one's promise is communicated using number of breaking one's promise is presented to other side, to cause huge obstruction to the social activity for object of breaking one's promise, can be effectively artificial at deterrence to breaking one's promise, and then promote it to correct discreditable behavior using the strength of public opinion.

Description

It breaks one's promise the display methods and device of number

Technical field

The present invention relates to field of computer technology, and in particular to a kind of display methods and device of number of breaking one's promise.

Background technique

Currently, China is accelerating to step into the credit epoch, personal credit problem has obtained more and more extensive concern.Accordingly Ground, personal credit start gradually mutually to link up with the every aspect in life.For example, certain medium platforms start to be believed according to individual Appraisal result provides a user infrastructure service, to bring huge convenience for user.

But at the same time still with the presence of the multiple discreditable behavior of part citizen, and plot is severe, has seriously affected society Meeting public order and the interests of the related common people.In existing way, for there are the citizens of multiple discreditable behavior, only when it Discreditable behavior offend to law baseline when, could law court in a manner of announcing and break one's promise roster to its personal information carry out It announces, to reach deterrent-effect.

But inventor has found in the implementation of the present invention, aforesaid way in the prior art at least exists following Defect：Since the roster that breaks one's promise that law court announces needs user actively to check and can know, and most people often has no time to check Such list, or even be unaware of this and go to check such list by which kind of channel.Therefore, lead to break one's promise people still in all kinds of social activities It is unblocked in activity, can not be veritably artificial at deterrence to breaking one's promise, it is even more impossible to promote it to correct discreditable behavior.It can be seen that Existing processing mode effect of breaking one's promise is undesirable.

Summary of the invention

In view of the above problems, it proposes on the present invention overcomes the above problem or at least be partially solved in order to provide one kind State the display methods and device of the number of breaking one's promise of problem.

According to an aspect of the invention, there is provided the display methods for number of breaking one's promise, including：To the target message got Word segmentation processing is carried out, feature vector corresponding with the target message is determined according to word segmentation processing result；

Described eigenvector is inputted into preset SMS classified model, when the output result of the SMS classified model is to lose When believing the result of type, communicating number corresponding with the target message is screened as destination number；

Judge whether the destination number is number of breaking one's promise, if so, for it is described break one's promise number setting in communication into The label of breaking one's promise of row display.

According to another aspect of the present invention, the display device for number of breaking one's promise is provided, including：

Word segmentation module, suitable for carrying out word segmentation processing to the target message got, according to the determination of word segmentation processing result and institute State the corresponding feature vector of target message；

Screening module, suitable for described eigenvector is inputted preset SMS classified model, when the SMS classified model Output result be break one's promise type result when, communicating number corresponding with the target message is screened as destination number；

Judgment module, suitable for judging whether the destination number is number of breaking one's promise, if so, being used for the number setting of breaking one's promise In the label of breaking one's promise shown in communication.

According to another aspect of the invention, a kind of electronic equipment is provided, including：Processor, memory, communication interface and Communication bus, processor, memory and communication interface complete mutual communication by communication bus；

Memory makes processor execute the aobvious of above-mentioned number of breaking one's promise for storing an at least executable instruction, executable instruction Show the corresponding operation of method.

In accordance with a further aspect of the present invention, a kind of computer storage medium is provided, at least one is stored in storage medium Executable instruction, executable instruction make processor execute the corresponding operation of display methods such as above-mentioned inactivating signal code.

The display methods and device of disclosed number of breaking one's promise according to the present invention, can be handled by the analysis to target message It determines number of breaking one's promise corresponding to object of breaking one's promise, and then is break one's promise mark of the number setting for being shown in communication of breaking one's promise Label.Which label that can will break one's promise when object of breaking one's promise is communicated using number of breaking one's promise is presented to other side, thus to breaking one's promise The social activity of object causes huge obstruction, can be effectively artificial at deterrence to breaking one's promise, and then utilizes the power of public opinion Amount promotes it to correct discreditable behavior.It can be seen that the processing mode effect of breaking one's promise in the present embodiment is even more ideal, be conducive to safeguard Social public order and the interests of the related common people.

The above description is only an overview of the technical scheme of the present invention, in order to better understand the technical means of the present invention, And it can be implemented in accordance with the contents of the specification, and in order to allow above and other objects of the present invention, feature and advantage can It is clearer and more comprehensible, the followings are specific embodiments of the present invention.

Detailed description of the invention

By reading the following detailed description of the preferred embodiment, various other advantages and benefits are common for this field Technical staff will become clear.The drawings are only for the purpose of illustrating a preferred embodiment, and is not considered as to the present invention Limitation.And throughout the drawings, the same reference numbers will be used to refer to the same parts.In the accompanying drawings：

Fig. 1 shows the flow diagram of the display methods of the number of breaking one's promise of one embodiment of the invention；

Fig. 2 shows the flow diagrams of the display methods of the number of breaking one's promise of another embodiment of the present invention；

Fig. 3 shows a kind of structure drawing of device of the display device of number of breaking one's promise of further embodiment of this invention offer；

Fig. 4 shows the structural schematic diagram of a kind of electronic equipment according to an embodiment of the present invention.

Specific embodiment

Exemplary embodiments of the present disclosure are described in more detail below with reference to accompanying drawings.Although showing the disclosure in attached drawing Exemplary embodiment, it being understood, however, that may be realized in various forms the disclosure without should be by embodiments set forth here It is limited.On the contrary, these embodiments are provided to facilitate a more thoroughly understanding of the present invention, and can be by the scope of the present disclosure It is fully disclosed to those skilled in the art.

Fig. 1 shows the flow diagram of the display methods of the number of breaking one's promise of one embodiment of the invention.As shown in Figure 1, This method includes：

Step S110：Word segmentation processing is carried out to the target message got, it is short with target according to the determination of word segmentation processing result Believe corresponding feature vector.

Wherein, target message can be whole short messages in network-wide basis, or be directed to specific user or specific interior The part of short message of appearance, which is not limited by the present invention.It can be determined by a variety of participle modes when word segmentation processing, as long as can obtain The each participle for including in target message.Feature vector corresponding with target message is determined according to word segmentation processing result When, corresponding feature vector directly can be determined according to the frequency of occurrence of each participle, it can also be determining pair using other modes The feature vector answered.It should for example, can be generated further combined with the default feature of other dimensions extracted from target message Feature vector.For another example, it can be combined with preset corpus word matrix and generate this feature vector.In short, the present invention for characteristic to The generating mode of amount without limitation, as long as being able to reflect the content characteristic of target message.

For example, in a kind of concrete implementation mode, it is contemplated that the malicious act that debtor falls into debt repeatedly often enrages Creditor, so that creditor is for indignation or threatens and sends the disagreeableness short message content of the tone to debtor.For this purpose, can be pre- Dictionary of breaking one's promise first is established, the dictionary of breaking one's promise is for storing the various vocabulary for abusing, threatening class.Correspondingly, preferentially by each participle It is matched with the vocabulary in dictionary of breaking one's promise, if it exists at least one participle to match with dictionary of breaking one's promise, then it can tentatively really The fixed target message may be related to type of dunning.

In addition, in a kind of concrete implementation mode term vector can also be trained in advance, and determine feature according to term vector Vector specifically when training, needs to screen the short message that transmission number is personal number from the sample short message being collected into and is trained. For example, stressing the short message for selecting personal mobile phone number to send to be trained, concrete reason is：Usually abuse the short message of property not Port can be walked to issue, therefore, by selecting the source of sample short message to be capable of the accuracy of effective training for promotion result.

Step S120：Feature vector is inputted into preset SMS classified model, when the output result of SMS classified model is When the result for type of breaking one's promise, communicating number corresponding with target message is screened as destination number.

Wherein, SMS classified model is broken one's promise the target message of type for identification.Specifically, SMS classified model can be straight It connected preset function or preset rules generates, can also be generated by modes such as machine learning and sample trainings, as long as energy Enough short messages for identifying type of breaking one's promise.When it is implemented, the short message conduct of multiple known types of breaking one's promise can be collected in advance Sample short message determines the feature vector of sample short message after then segmenting to sample short message, and then according to the spy of sample short message Sign vector training obtains SMS classified model.Wherein, the generating mode of the feature vector of sample short message can directly reference object it is short The generating mode of the feature vector of letter.

Step S130：Judge whether destination number is number of breaking one's promise, if so, for break one's promise number setting for communication when into The label of breaking one's promise of row display.

It specifically, can be according to preset judgement precision and applied field when whether judge destination number is to break one's promise number Scape determines specific judgment mode.For example, when application scenarios are more wide in range, and when judging that precision is lower, it can be directly by target Number is determined as number of breaking one's promise；It for another example, can be according to preset judgement when application scenarios are more harsh, and judge that precision is higher Rule judges whether destination number is number of breaking one's promise.Specific judgment rule either carry out assistant analysis judgement by cloud, Be also possible to by client carry out aid mark judgement, in short, the present invention to specific judgment mode without limitation.

When determining destination number is to break one's promise number, break one's promise to break one's promise number setting for what is shown in communication Label.Specifically, it can be stored in Cloud Server side and break one's promise number and its corresponding label of breaking one's promise correspondingly is monitored when subsequent When communication event relevant to the number of breaking one's promise, this is broken one's promise into label display in client corresponding with the communication event.Or Person, Cloud Server side can also be stored break one's promise number and its corresponding label of breaking one's promise is handed down to each client, for each A client when detecting communication event relevant to the number of breaking one's promise, by this break one's promise label display with the communication event pair In the client answered.

In short, above-mentioned each step can both be realized by client-side, can also be realized by server side, it can also be by Server and client side realizes jointly.But since server can get the target message of magnanimity, it is preferred, therefore, that right The process that target message is screened is realized in Cloud Server side.

The display methods of disclosed number of breaking one's promise according to the present invention can be handled by the analysis to target message and determine mistake Believe number of breaking one's promise corresponding to object, and then is break one's promise label of the number setting for being shown in communication of breaking one's promise.The party The formula label that can will break one's promise when object of breaking one's promise is communicated using number of breaking one's promise is presented to other side, thus to the society for object of breaking one's promise It turns over a finished item to move and causes huge obstruction, can be effectively artificial at deterrence to breaking one's promise, and then promote it using the strength of public opinion Correct discreditable behavior.It can be seen that the processing mode effect of breaking one's promise in the present embodiment is even more ideal, it is public to be conducive to maintenance society Order and the interests of the related common people.

Fig. 2 shows the flow diagrams of the display methods of the number of breaking one's promise of another embodiment of the present invention.Such as Fig. 2 institute Show, this method includes：

Step S200：Word segmentation processing is carried out to the sample short message got, it is short with sample according to the determination of word segmentation processing result Corresponding feature vector is believed, according to the SMS classified model of feature vector corresponding with sample short message training.

Wherein, the sample short message in the present embodiment includes：It duns short message, law court or lawyer's letter's short message, and/or credit platform All kinds of short messages such as short message.Sample short message can be obtained by modes such as server active collection and/or user feedbacks.Correspondingly, originally Number of breaking one's promise in embodiment includes：Lao Lai number (number for the user not paid the money one owes).Firstly, for the ease of training, it can Sample short message is marked be positive sample or negative sample in advance.Wherein, positive sample is the sample of type of breaking one's promise, and negative sample is non-mistake Believe the sample of type.Optionally, positive sample can also be further labeled as to the positive sample of multiple grades.For example, in this implementation In example, positive sample can be divided by three grades according to the source of positive sample, wherein the positive sample of the first estate mainly includes The excited short message of dunning that user sends, the positive sample of the second grade mainly include from law court or rule official document class it is short Letter, the positive sample of the tertiary gradient mainly includes the delay refund short message from credit card platform or loan platform.Specifically, it is drawing When graduation, a kind of (positive sample of i.e. above-mentioned the first estate), phase are classified as by the contents such as be related to personal attack, abuse, threaten Ying Di needs that dictionary mentioned above of breaking one's promise is combined to determine whether short message sample belongs to the positive sample of the first estate；It will be related to drawing The content dunned with legal provision is classified as a kind of (positive sample of i.e. above-mentioned second grade), correspondingly, can build in advance It makes laws a dictionary, for storing legal provision relevant with the affairs such as debt repayment, and the combination law article dictionary determines short message sample Whether this belongs to the positive sample of the second grade；It will be related to that the loan platforms such as p2p, credit card are overdue etc. normal the returning for contents such as to remind For a kind of (positive sample of the i.e. above-mentioned tertiary gradient), correspondingly, credit dictionary can be pre-established, for storage and character loan Relevant wording, and determine whether short message sample belongs to the positive sample of the tertiary gradient in conjunction with the credit dictionary.Then, after to mark Sample short message carry out word segmentation processing, and corresponding with sample short message feature vector is determined according to word segmentation processing result.Wherein, The corresponding feature vector of sample short message includes：Short message term vector, theme vector, and/or numeric type vector etc..At participle Complicated operation for reason and determining feature vector, therefore, for the compactedness of description, the contents of the section is placed on this implementation The end of example, which is concentrated, to be illustrated, details are not described herein again.Finally, when according to the feature vector SMS classified model of training, it can be by a variety of Machine learning algorithm is trained, for example, can be trained by neural network scheduling algorithm.Also, it can also be according to user's The disaggregated model that the factors such as feedback obtain training is modified, with lift scheme precision.

Step S210：Word segmentation processing is carried out to the target message got, it is short with target according to the determination of word segmentation processing result Believe corresponding feature vector.

Wherein, target message includes being sent to number of breaking one's promise, all kinds of short messages for dunning.The present embodiment can be to whole Short message be monitored, can also be monitored according to factors such as occupation, the credits of number owner and meet the short message of preset condition, example Such as, potential creditor, law practitioner or short message of credit platform etc. are only monitored.In addition, due to word segmentation processing and determining feature Complicated operation for vector, therefore, for the compactedness of description, the end that the contents of the section is placed on the present embodiment is concentrated and is explained It states, details are not described herein again.

Step S220：Feature vector corresponding with target message is inputted into preset SMS classified model, when short message point The output result of class model be break one's promise type result when, communicating number corresponding with target message is screened as target number Code.

The result for type of breaking one's promise further is divided by three grades by SMS classified model in this present embodiment, thus For determining multiple and different grades of breaking one's promise, therefore, when determine output result be break one's promise type result when, further judge defeated The corresponding grade of breaking one's promise of result out, and determine according to the corresponding grade of breaking one's promise of output result the class information of destination number.

It, may be related to number of breaking one's promise since target message is large number of, it is also possible to it is unrelated with number of breaking one's promise, therefore, The essence of this step is equivalent to according to SMS classified model judge whether target message is related to number of breaking one's promise, if so, will be with The corresponding communicating number screening of target message is destination number.Wherein, communicating number corresponding with target message includes：Mesh Mark the number for including in the number of recipient, the number of sender of target message and the text of target message of short message.Tool When body is realized, the concrete type of communicating number corresponding with target message can be judged according to the content of target message.Than Such as, when the content of target message is to dun content, then the number of the recipient of target message can be determined as destination number； For another example, when the content of target message is the content for refusing to pay one's debt, then the number of the sender of target message can be determined as Destination number；For another example, when the content of target message is that the subpoena that issues of law court notifies what class short message or credit platform issued to press for payment of When refund short message, it may include all kinds of identity informations such as the number information of accommodator in the text of target message, then can incite somebody to action The number for including in the text of target message is determined as destination number.

Step S230：Judge whether destination number is number of breaking one's promise.

Screened in previous step in order to prevent destination number inaccuracy, promote the accuracy of identification, by this step into One step judges whether destination number is number of breaking one's promise.When specifically judging, various ways can be used, mainly use visitor in the present embodiment The mode of family end aid mark：When monitoring communication event relevant to destination number, in corresponding terminal circle of communication event Show in face for prompting user break one's promise the prompt information of class label processing for the destination number；According to receive with The corresponding confirmation flag information of prompt information determines whether destination number is number of breaking one's promise.Specifically, first destination number is added Enter in preset monitoring list, which both can store can also be handed down to each client on Cloud Server, with Just communication event relevant to destination number is monitored in the follow-up process.Wherein, communication event relevant to destination number includes： Calling event goes electric event, short message event, the chat event based on all kinds of chat applications etc..In short, all joined by destination number It can be used as the monitoring object in the present invention with the event of communication.When monitoring communication event relevant to destination number, Show in the corresponding terminal interface of communication event for prompting user break one's promise the mentioning of class label processing for the destination number When showing information, which is similar to forms such as " user may be object of breaking one's promise, PLSCONFM ".In addition, for prompting to use Family break one's promise for the destination number class label processing prompt information in can also further include the grade of destination number Information, for example, the prompt information is " user may be that level-one is broken one's promise object, PLSCONFM " etc..The class information of destination number Guidance can be provided for the confirmation operation of user.It, can be in call circle of client when communication event is incoming call or goes electric event Prompt information is added in face, and the entrance for confirmation is set, so that user sends confirmation flag information；When communication event is short When believing class event, the short message comprising prompt information can be issued to client, and comprising true for guiding user to send in short message Recognize the content of mark information.

In addition, determining whether destination number is mistake according to the confirmation flag information corresponding with prompt information received When signal code, judge by accident in order to prevent, it can be using the memory judgement of at least one of following three kinds of modes：

In the first way, determine whether the number of the confirmation flag information corresponding with destination number received is big In preset threshold, determine whether destination number is number of breaking one's promise according to judging result.For example, if receive with destination number phase The number of corresponding confirmation flag information is more, then illustrate the number be break one's promise number probability it is larger, correspondingly, determined For number of breaking one's promise.

In the second way, terminal corresponding to the confirmation flag information corresponding with destination number received is obtained The user identifier of user determines whether destination number is number of breaking one's promise according to the number of user identifier.Wherein, user identifier can The information such as one user of unique identification, such as the cell-phone number that can be user.It correspondingly, can be true according to the number of user identifier Whether the fixed user that the destination number is marked is the same user, if the same user, even if there are many label number, But, it would still be possible to it is cheating, therefore, only just determines that destination number is inactivating signal when the number of user identifier is more Code.

In the third mode, the time corresponding to the confirmation flag information corresponding with number of breaking one's promise received is obtained Information determines whether destination number is number of breaking one's promise according to the period distribution situation of temporal information.Corresponding to confirmation flag information Temporal information refer to that user sends the time of the confirmation flag information, corresponding to the confirmation mark information that each time is received Temporal information is drawn on a timeline, so that it is determined that the period distribution situation of temporal information, if each secondary confirmation flag information collects In in the same period then illustrate that there may be same users to pass through the case where multiple equipment is practised fraud simultaneously；Conversely, if each confirmation Period locating for mark information it is more dispersed and it is irregular say, then illustrate the number be break one's promise number a possibility that it is bigger.

In actual conditions, above-mentioned three kinds of modes can be used in combination, to further promote judging nicety rate.

In addition, can also be realized further combined with other modes, example when whether judge destination number is to break one's promise number Such as, by preset law court's class official information storage into official's database, in conjunction with official's database judge destination number whether be It breaks one's promise number.Wherein, official's database is used to store is determined as breaking one's promise the list list of object by law court.

Step S240：If so, for break one's promise label of the number setting for being shown in communication of breaking one's promise.

Wherein, which can be used various ways realization, for example, can be directly by the number and corresponding of breaking one's promise Label of breaking one's promise is supplied to corresponding operator, and corresponding mark of breaking one's promise is arranged for the number of breaking one's promise by the communication server by operator Label, to show the label of breaking one's promise to corresponding terminal user when the number of breaking one's promise carries out any communication behavior.For another example, may be used This is directly broken one's promise number and corresponding label of breaking one's promise is stored on Cloud Server, correspondingly, when terminal device execute it is logical When letter event, first passes through the client software installed on terminal device and send request to Cloud Server, to be sentenced by Cloud Server Whether the number to break in the communication event is number of breaking one's promise, if so, notice client software shows corresponding label of breaking one's promise.

When specific implementation, will break one's promise number and label associated storage of breaking one's promise corresponding with number of breaking one's promise are to Cloud Server In, to create and use in communication terminal corresponding with number of breaking one's promise when to monitor communication event relevant to number of breaking one's promise In the display interface for showing label of breaking one's promise.

It optionally, can also be further by break one's promise number and mistake corresponding with number of breaking one's promise stored in cloud server Beacon, which is signed, issues communication terminal corresponding with number of breaking one's promise, and is saved so that communication terminal is locally downloading, so as to again When monitoring communication event relevant to number of breaking one's promise, according to the number and corresponding with number of breaking one's promise of breaking one's promise locally saved Label of breaking one's promise creation is for showing the display interface for label of breaking one's promise.When specifically issuing, it can break one's promise what is saved in Cloud Server Number and its label of breaking one's promise uniformly are stored in record sheet of breaking one's promise, so that the full content in the record sheet of breaking one's promise be issued respectively To each communication terminal.Optionally, in order to save the memory space of each communication terminal, each communication can also be analyzed in advance eventually Contact information (i.e. address list) on end, and then the pass with the communication terminal is issued to each communication terminal based on the analysis results The biggish number of breaking one's promise of connection property, to promote the specific aim issued.Wherein, with the biggish inactivating signal of the relevance of the communication terminal Code is determined according to the minimal path value between each node in preset contact person's chart.For example, can be according to each communication Contact information in terminal constructs contact person's chart, which is made of node and side, wherein by each contact person Corresponding communication identifier (such as communicating number) is used as a node, is identified with side and is directly associated between two nodes.For example, It is related in the cell phone address book of contact person A (corresponding communication identifier A) number of people B (corresponding communication identifier B), alternatively, communication There is history message registration of certain time length etc. between mark A and communication identifier B, then in contact person chart interior joint A and section Establishing between point B has a line.Correspondingly, for the user of each communication terminal, the corresponding each section of the user is analyzed Minimal path value is less than the node of certain threshold value as relevant to the user by the minimal path value between point and other nodes Node, and then the number of breaking one's promise for including in node relevant to the user is only sent to the user.Wherein it is possible to pass through following Formula calculates the minimal path value between two nodes (first node and second node) either in two ways：(1) it calculates Minimal path value between first node and second node, including：First node and the second section are calculated using shortest path first Shortest path between point；Using the number on the side for including in shortest path as minimal path value.(2) first node and the are calculated Minimal path value between two nodes, including：Obtain the All Paths between first node and second node；To first node and Either path between second node obtains the weighted value on each side on the path, calculates the weighted value on all sides in the path The sum of be used as weighted sum；Wherein, on the path it is each while weighted value and this while two node corresponding relationship sides between go through The history number of calls has negatively correlated incidence relation；Weighted sum in All Paths between first node and second node is minimum Path as shortest path, using corresponding weighted sum as minimal path value.Furthermore it is also possible to according to the occupation of terminal user The all types of user such as information, regional information attribute information issues and the biggish mistake of the relevance of the communication terminal to each communication terminal Signal code.

Wherein, above-mentioned label of breaking one's promise can be used for the end of the communication corresponding to all kinds of communication events relevant to number of breaking one's promise Show in end.Wherein, all kinds of communication events relevant to number of breaking one's promise include：By incoming call thing of the number as calling party of breaking one's promise Part, by break one's promise number as callee go electric event, by break one's promise number as sender or recipient short message event or chat Its event etc..It can be seen that the number of breaking one's promise for object of breaking one's promise can be screened based on short message through the above way, and then in its communication When show and break one's promise label to communication participant.The necessary means of social development in today's society are had become due to communicating, lead to It crosses and shows label of breaking one's promise in communication event, the affiliate for the object that can make to break one's promise understands its discreditable behavior, and then this is made to break one's promise The social process of object is obstructed, so that it be forced quickly to stop discreditable behavior (such as cash payment on account under the pressure of public opinion Deng).

In addition, finally elaborating word segmentation processing and determining the operating process of feature vector.It is with target message below Example is described, and the operating process of sample short message is similar therewith, and details are not described herein again.Specifically, in word segmentation processing, in addition to adopting It is segmented, can also be segmented using participles tools such as IKAnalyzer, Pan Gu's participles, the embodiment of the present application is not limited with Ali Segment the specific tool used.After participle, need further to determine feature vector according to word segmentation result, wherein feature vector Generating mode include at least one of following four：

In the first way, feature vector includes short message term vector.Specifically, word segmentation processing is carried out to target message, To obtain at least one participle of target message；Respectively by least one segment in each participle input term vector tool obtain At least one term vector；By all term vectors and the acquisition of preset corpus word matrix multiple and target at least one term vector The corresponding short message term vector of short message.It can be seen which after participle, further obtains the term vector of each participle, lead to Term vector is crossed to indicate the semanteme of each participle.Term vector is that each of language word is mapped to a fixation by training The vector of length can obtain the term vector of each participle by the way that each participle is inputted term vector tool, such as can pass through Word2vec tool obtains each participle as Feature Mapping to K dimension space the term vector of participle.Wherein, in term vector tool Include corpus word matrix, be the matrix of n*k in the corpus word matrix, n is the participle amount in database, and k is the dimension of each word Degree, it is assumed that comprising 800 participles, the dimension of term vector set as 100 in a certain term vector tool, then the corpus word matrix It is then the matrix of 800*100.Will participle input term vector tool, term vector tool can obtain one indicate the participle position to It measures and this vector and corpus word matrix multiple can be obtained and export the term vector of this participle.Obtaining target message After the term vector of each participle, further, in order to indicate the semanteme of target message, by by all participles of target message Term vector product and corpus word matrix multiple obtain target message target term vector.Such as：Assuming that the corpus word matrix For the matrix T of 800*100, if target message has 10 participles, n1~n10 respectively indicates the term vector of this 10 participles, will The product and corpus word matrix T-phase of n1*n2* ... n10 multiply the target term vector for obtaining target message, i.e. target term vector is equal to n1*n2*…n10*T。

In the second way, feature vector includes theme vector.Specifically, the target message got is segmented Processing segments when determining feature vector corresponding with target message according to word segmentation processing result according to target message At least one participle obtained after processing and preset semantic topic generate model and obtain theme corresponding with target message Vector.For example, it is short to obtain target by LDA (Latent Dirichlet Allocation, semantic topic generate model) Theme vector, that is, theme belonging to letter.Specifically, all participles at least one participle of target message are inputted LDA mould Type, LDA model can correspond to each participle and obtain the vector for indicating its affiliated theme topic probability, then by all points of input The addition of vectors of its affiliated theme topic probability of the expression of word, in the theme matrix phase of the vector that will add up and LDA model Multiply the theme vector for obtaining target message and output.Such as：Assuming that the theme matrix of LDA model is the matrix T1 of 10*100, if Target message has 10 participles, and m1~m10 respectively indicates this 10 vectors for segmenting affiliated theme topic probability, by m1* M2* ... m10, which is added then to be multiplied with theme matrix T1 again, obtains the theme vector of target message, the i.e. theme vector of target message Equal to (m1+m2+ ... m10) * T1.Preferably, the second way can be combined with first way, for example, by least one Each participle in participle is combined as second feature, theme vector as third feature as fisrt feature, short message term vector At an eigenmatrix (being equivalent to feature vector mentioned above), by the SMS classified model of this feature Input matrix to target message Classify, to determine whether it is the short message for type of breaking one's promise.Correspondingly, SMS classified model obtains by the following method：It adopts With preset SMS classified rule, construct polytypic SMS classified model, wherein parameters in SMS classified model according to It is configured according to the attributive character of short message, attributive character includes the participle, term vector and theme vector of short message；Obtain plurality of classes Short message, wherein every other short message of type include it is a plurality of；Using the short message of plurality of classes as training sample to SMS classified model It is trained, calculates the parameters in SMS classified model, the SMS classified model after being trained.Which is according to short message Participle, term vector and theme vector classify, increase classification can foundation feature, enable the classification to short message more It is careful, deep, improve the accuracy of result.In addition, since term vector not only characterizes the semanteme of short message itself, additionally it is possible to characterize Relationship between short message and similar short message, and theme vector can characterize theme belonging to short message and between other themes Relationship can help SMS classified model to divide new, similar short message by the term vector and theme vector of short message Class improves the generalization ability of SMS classified model, that is, improves the resolution to short message.

In the third mode, feature vector includes numeric type vector.Specifically, the target message got is divided Word is handled, and when determining feature vector corresponding with target message according to word segmentation processing result, includes in acquisition target message Each participle and digital information calculate separately the distance that each participle arrives each digital information on the basis of digital information；It will Distance is used as weight, and each participle is used as dimension, is described short message content to generate number corresponding with target message Type vector.Due to that in actual conditions, can include digital information in many short messages, for example, consumption information, remaining sum that bank sends be believed Breath or the short breath for pressing for payment of credit balance sent comprising the refund short message of repayment amount and law court sent of credit platform with And the short message etc. comprising amount owed that creditor sends.Therefore, it is necessary to first parse the digital information in short message, and for number Information is targetedly handled.Parsed again after specifically can first deleting the specific information in original short message, to prevent major structure at Interference.Wherein, specific information includes the information such as URL, the IP address in short message, by those information deletions, to prevent from disappearing to characterization Take the interference of the useful digital informations such as the amount of money, logistics odd numbers.In short, since digital information is for text information, it may Significance level is higher, larger for the judgement meaning of amount owed and telephone number in short message of dunning, therefore, passes through extraction Digital content in short message can preferably promote the accuracy of judging result to generate numeric type vector.When specific implementation, also Predict the meaning for each digital representation for including in short message, in combination with condition random field algorithm to analyze whether it is to break one's promise Number or debt information.

The generating mode of above-mentioned several feature vectors can be used alone, and can also be used in combination, when used in combination, Each type of vector respectively as feature vector a dimension, thus generate a multidimensional characteristic vectors.Correspondingly, short message The training process of disaggregated model and the generating mode of feature vector are corresponding.By the above-mentioned means, can accurately determine target The type of short message to filter out the short message for type of breaking one's promise, and analyzes corresponding number of breaking one's promise accordingly.

In addition, in above process, it is contemplated that the malicious act that debtor falls into debt repeatedly often enrages creditor, so that Creditor is for indignation or threatens and sends the disagreeableness short message content of the tone to debtor.For this purpose, mistake can also be pre-established Believe dictionary, the dictionary of breaking one's promise is for storing the various vocabulary for abusing, threatening class.Correspondingly, preferentially by each participle and word of breaking one's promise Vocabulary in allusion quotation is matched, at least one participle to match with dictionary of breaking one's promise, then can primarily determine the target if it exists Short message may be related to type of dunning.Further, can also according to the quantity of participle to match with dictionary of breaking one's promise and Determine that target message is the probability of class short message of dunning with degree score.

In addition, in above process, term vector can be trained in advance, term vector mentioned above is obtained according to training result Tool.When specific training, needs to screen the short message that transmission number is personal number from the sample short message being collected into and be trained. For example, stressing the short message for selecting personal mobile phone number to send to be trained, concrete reason is：Usually abuse the short message of property not Port can be walked to issue, therefore, by selecting the source of sample short message to be capable of the accuracy of effective training for promotion result.

In addition, above-mentioned analysis short message content and generating the operation of feature vector and preferably realizing beyond the clouds.Correspondingly, realizing When, the operating system in communication terminal device can be generally the open some interfaces of application program, obtain from system for application program Take required information.Therefore, in embodiments of the present invention, it is monitored in order to the short message reception condition to communication terminal, An application program can be run in communication terminal, then can utilize this feature of communication terminal system, be in operation Application program is registered in system.In this way, when operating system broadcast notice messages, application program can uppick it is this Notification message, and therefrom obtain required information.For example, the operating system of communication terminal will broadcast this when receiving short message Short message receives message, and carries the telephone number of SMS sender, in this way, application program can monitor the event, and from In get the telephone number of SMS sender.In short, the short message phase that client (i.e. communication terminal device) can will acquire Information reporting is closed to Cloud Server, for Cloud Server analysis.Optionally, short message complete content is reported to cloud in order to prevent The phenomenon that damaging privacy of user occurs, and can carry out default place to the short message relevant information that client is got in the present embodiment Reason, so that Cloud Server can not get original short message content.This presets processing：Hash processing etc..For example, working as client When termination receives short message, the corresponding cryptographic Hash of content of short message is calculated；The corresponding cryptographic Hash of the content of short message is uploaded into cloud Server is held, so that cloud server classifies to short message according to the cryptographic Hash.Wherein, client calculates the content of short message The process of corresponding cryptographic Hash specifically includes：Word segmentation processing is carried out to the content of short message；Each word after participle is assigned different Vector value, summarize the corresponding simhash value of content that short message is calculated.

In summary, mode provided in this embodiment can effectively excavate potential inactivating signal using the short message being collected into Code, and the modes such as user annotation is combined to ensure to break one's promise the accuracy of number.Moreover, by being arranged for number of breaking one's promise for communicating When the label of breaking one's promise that shows, the label of breaking one's promise can be presented to this communication when the number of breaking one's promise carries out all kinds of communication behaviors Stakeholder, to achieve the effect that promote to break one's promise by way of public opinion and common people's supervision, people fulfils obligations.Due to The daily communication of people be unable to do without communication behavior, and therefore, the deterrent-effect of the method is ideal.

In addition, those skilled in the art can also carry out various improvement and deformations to aforesaid way.For example, short in order to save Believe the training time of disaggregated model, promotes the training effectiveness of sample short message, can use existing for realizing in other scenes The models of various classification carry out transfer learning, to obtain the SMS classified model for the application scenarios being suitable for of the invention.Its In, the existing model for realizing the various classification in other scenes is referred to as old model, which is by magnanimity sample What is generated after this training can identify the model of thousands of kinds of classification.I.e.：The applicable scene of old model is wide in range, classification is many kinds of. But if old model directly applies in the identification scene of the short message in the application, it can cause to know due to not having specific aim The accuracy of other result is too low.To solve the above-mentioned problems, the present embodiment passes through the deep learning that has obtained in old model It is superimposed the mode of the fully-connected network of multilayer on network, it is (i.e. SMS classified in the application to obtain the stronger new model of specific aim Model).New model can finally export 6-8 classification needed for short message content identification.It can be seen that by be newly collected into The relevant sample short message of type of breaking one's promise carries out the mode of transfer learning to old model, can be obviously improved the specific aim of new model, It is set to take into account accuracy and training effectiveness.Specifically, retain the parameter of all convolutional layers in trained old model, and will be old The full articulamentum of the last layer in model replaces with the full Connection Neural Network an of multilayer.Because trained old model is Through having the understandability to short message content, so the huge parameter of nervous layer in addition to the last layer does not need to instruct again Practice, it is only necessary to which thus training effectiveness is substantially improved in the newly added full Connection Neural Network of training.Full articulamentum is generally by two Part forms：A part is linear segment, mainly does Linear Mapping；Another part is non-linear partial, for linear transformation Output result do non-linear conversion.It can be seen which can be identified using the sample information of breaking one's promise got to first Model (i.e. above-mentioned old model) carries out transfer learning, to obtain the second identification model (i.e. above-mentioned new model).

In addition, in the present embodiment, in order to which the exhibition method for the number that makes to break one's promise is more flexible, understand also for convenient for the public The concrete condition of corresponding executed person of breaking one's promise may further determine that the number attribute information and/or associated diagram of number of breaking one's promise As information.Wherein, the number attribute information of number of breaking one's promise and/or associated images information can be according in the communication interfaces of terminal device Preset Communication Review entrance obtains.For example, after detecting communication event, corresponding with the communication event logical Believe and pop up Communication Review entrance in interface, evaluated for the communication object in this communication event for user's input in Hold, age, the occupational information that can specifically input the information of breaking one's promise about communication object, can also input communication object, it might even be possible to The photo of uplink object, video information.Cloud Server receives what each communication terminal was arrived by Communication Review inlet collector About the comment information of each communication object, and according to above-mentioned comment information determine break one's promise number number attribute information and/or Associated images information.When it is implemented, the Communication Review entrance can be open only for the terminal user for participating in communication event, example Such as, after user A and user B terminate call, user A and user B can for the communication object in this communication event into Row comment.User A can both comment on user B (such as complain, report, thumb up), and user A can also carry out itself It comments on (such as occupation for uploading the image of itself, inputting itself, to realize propaganda objective).User's malice is commented in order to prevent By, the terminal user for having neither part nor lot in communication event can not comment on for the communication object in corresponding communication event, for example, with This user C for conversing unrelated can not participate in commenting on.Correspondingly, in mistake of the number setting for being shown in communication of breaking one's promise When beacon label, the tag attributes letter for including in the label of breaking one's promise of the number of breaking one's promise is determined according to the number attribute information for number of breaking one's promise Breath, and/or, the label image letter for including in the label of breaking one's promise of the number of breaking one's promise is determined according to the associated images information for number of breaking one's promise Breath.Wherein, label attribute information includes at least one of the following：Information for the grade that indicates to break one's promise (such as can be according to mistake The severity of letter divides multiple grades of breaking one's promise), for the time that indicates to break one's promise information (such as indicate specific number of days of breaking one's promise, Number of days is longer to illustrate that its behavior is more severe), for the event that indicates to break one's promise information (such as indicate specific event type, it is such as deficient Money class refuses reparation class, does not abide by public regulation class) and for the object identity that indicates to break one's promise information (such as object of breaking one's promise Identification card number, post, native place etc.).The particular row for object of breaking one's promise can be understood convenient for other users by label attribute information For and identity.Label image information includes at least one of the following：Information corresponding with the photo for object of breaking one's promise or video (the personal photo for executed person of such as breaking one's promise) and information corresponding with the photo for the event of breaking one's promise or video (such as event pair The video recording etc. answered).The macroscopic features that object of breaking one's promise can be understood convenient for other users by label image information, thus to breaking one's promise Object causes to deter.In addition, when label is broken one's promise in display, it can be preset and multiple correspond respectively to different label attribute informations And/or the indicating template of breaking one's promise of label image information, from the label that determines and break one's promise in preset multiple indicating templates of breaking one's promise The indicating template of breaking one's promise that label attribute information and/or label image information match；Basis and label of breaking one's promise in terminal device In the indicating template of breaking one's promise that matches of label attribute information and/or label image information show label of breaking one's promise.For example, can distinguish For the label of breaking one's promise of different grades of breaking one's promise, different indicating templates of breaking one's promise is set：Higher grade of breaking one's promise, it is corresponding break one's promise it is aobvious Show that the quantity for the template elements for including in template is abundanter, in favor of showing its personal information comprehensively, reaches stronger deterrence effect Fruit.For another example, it can be directed to the label of breaking one's promise of different types of label image information respectively, different indicating templates of breaking one's promise is set：Mark The class template that corresponds to that image information is static images is signed, label image information is dynamic video corresponding to another kind of mould Plate, in order to more preferably show effect to realize according to the size that each template elements for including in template are arranged the characteristics of image Fruit.When it is implemented, shown according to indicating template of breaking one's promise when breaking one's promise label, firstly, determining in the indicating template of breaking one's promise and including The type and quantity of template elements；Then, the element value of each template elements is determined respectively, and the element value is for indicating corresponding element The information such as position, size, shape and the display content of element；Finally, respectively with the corresponding fills of the element value of each element The respective element broken one's promise in indicating template, to generate label of breaking one's promise.Specifically, terminal device can be shown by way of floating layer It breaks one's promise label, and the transparency of floating layer can also be controlled, to improve visual effect.Wherein, which can be by terminal Equipment is locally carrying out drafting generation, and correspondingly, terminal device locally presets the correlative code for generating label of breaking one's promise.Or Person, the label of breaking one's promise can also be pre-rendered by Cloud Server and be generated, and correspondingly, terminal device only needs to break one's promise what is received Label is presented.

Fig. 3 shows a kind of apparatus structure of the display device of number of breaking one's promise of further embodiment of this invention offer Figure, as shown in figure 3, described device includes：

Word segmentation module 31, suitable for carrying out word segmentation processing to the target message that gets, according to word segmentation processing result determine with The corresponding feature vector of the target message；

Screening module 32, suitable for described eigenvector is inputted preset SMS classified model, when the SMS classified mould The output result of type be break one's promise type result when, communicating number corresponding with the target message is screened as target number Code；

Judgment module 33, suitable for judging whether the destination number is number of breaking one's promise, if so, for the number setting of breaking one's promise Label of breaking one's promise for being shown in communication.

Optionally, the judgment module 33 is particularly adapted to：When monitoring communication event relevant to the destination number, Show in the corresponding terminal interface of the communication event for prompting user to carry out class mark of breaking one's promise for the destination number The prompt information of reason；According to the confirmation flag information corresponding with the prompt information received, the destination number is determined It whether is number of breaking one's promise；

Wherein, the communication event includes：Calling event and/or go electric event.

Optionally, the SMS classified model is for determining multiple and different grades of breaking one's promise；When the SMS classified model Output result be break one's promise type result when, the screening module 32 is further adapted for：Judge that the output result is corresponding It breaks one's promise grade, the class information of the destination number is determined according to the corresponding grade of breaking one's promise of the output result；

And it is described for prompt user for the destination number break one's promise class label processing prompt information in further Class information comprising the destination number.

Optionally, the judgment module 33 is particularly adapted to：

Determine whether the number of the confirmation flag information corresponding with the destination number received is greater than preset threshold, Determine whether destination number is number of breaking one's promise according to judging result；And/or

Obtain the user of terminal user corresponding to the confirmation flag information corresponding with the destination number received Mark, determines whether the destination number is number of breaking one's promise according to the number of the user identifier；And/or

Temporal information corresponding to the confirmation flag information corresponding with the number of breaking one's promise received is obtained, according to institute The period distribution situation for stating temporal information determines whether the destination number is number of breaking one's promise.

Optionally, the judgment module 33 is particularly adapted to：

By it is described break one's promise number and it is corresponding with the number of breaking one's promise break one's promise label associated storage into Cloud Server, To be created in communication terminal corresponding with the number of breaking one's promise when monitoring communication event relevant to the number of breaking one's promise Build the display interface for showing the label of breaking one's promise.

Optionally, the judgment module 33 is further adapted for：

By break one's promise number and the label of breaking one's promise corresponding with the number of breaking one's promise stored in cloud server It is handed down to the communication terminal corresponding with the number of breaking one's promise, is saved so that the communication terminal is locally downloading, with When just monitoring communication event relevant to the number of breaking one's promise again, break one's promise according to locally saving number and with institute The corresponding label creation of breaking one's promise of number of breaking one's promise is stated for showing the display interface of the label of breaking one's promise.

Optionally, the judgment module 33 is further adapted for：

By the storage of preset law court's class official information into official's database, the mesh is judged in conjunction with official's database Whether label code is number of breaking one's promise.

Optionally, described eigenvector includes：Short message term vector, then the word segmentation module 31 is particularly adapted to：

Word segmentation processing is carried out to the target message, to obtain at least one participle of the target message；

Each participle input term vector tool at least one described participle is obtained at least one term vector respectively；

By all term vectors and the acquisition of preset corpus word matrix multiple and the mesh at least one described term vector Mark the corresponding short message term vector of short message.

Optionally, described eigenvector includes：Theme vector, then the word segmentation module 31 is particularly adapted to：

According at least one participle and preset semantic topic for obtain after word segmentation processing to the target message It generates model and obtains theme vector corresponding with the target message.

Optionally, described eigenvector includes：Numeric type vector, then the word segmentation module 31 is particularly adapted to：

The each participle for including in the target message and digital information are obtained, on the basis of the digital information respectively Calculate the distance that each participle arrives each digital information；

Using the distance as weight, each participle is used as dimension, is described short message content to generate and the mesh Mark the corresponding numeric type vector of short message.

Optionally, described device further comprises：Training module 34；And the training module 34 is particularly adapted to：

Word segmentation processing is carried out to the sample short message got, it is opposite with the sample short message according to the determination of word segmentation processing result The feature vector answered；

According to feature vector corresponding with the sample short message training SMS classified model；

Wherein, the corresponding feature vector of the sample short message includes：Short message term vector, theme vector, and/or numeric type Vector.

Optionally, the sample short message includes：Short message, law court or the lawyer's letter's short message of dunning, and/or credit platform short message, The number of breaking one's promise includes：Lao Lai number.

The specific structure and working principle of above-mentioned modules can refer to the description of corresponding steps in embodiment of the method, herein It repeats no more.

The embodiment of the present application provides a kind of nonvolatile computer storage media, and computer storage medium is stored at least The display side of the number of breaking one's promise in above-mentioned any means embodiment can be performed in one executable instruction, the computer executable instructions Method.

Fig. 4 shows the structural schematic diagram of a kind of electronic equipment according to an embodiment of the present invention, the specific embodiment of the invention The specific implementation of electronic equipment is not limited.

As shown in figure 4, the electronic equipment may include：Processor (processor) 402, communication interface (Communications Interface) 404, memory (memory) 406 and communication bus 408.

Wherein：

Processor 402, communication interface 404 and memory 406 complete mutual communication by communication bus 408.

Communication interface 404, for being communicated with the network element of other equipment such as client or other servers etc..

Processor 402, for executing program 410, in the display methods embodiment that can specifically execute above-mentioned number of breaking one's promise Correlation step.

Specifically, program 410 may include program code, which includes computer operation instruction.

Processor 402 may be central processor CPU or specific integrated circuit ASIC (Application Specific Integrated Circuit), or be arranged to implement the integrated electricity of one or more of the embodiment of the present invention Road.The one or more processors that electronic equipment includes can be same type of processor, such as one or more CPU；It can also To be different types of processor, such as one or more CPU and one or more ASIC.

Memory 406, for storing program 410.Memory 406 may include high speed RAM memory, it is also possible to further include Nonvolatile memory (non-volatile memory), for example, at least a magnetic disk storage.

Program 410 specifically can be used for so that processor 402 executes the number of breaking one's promise in above-mentioned any means embodiment Display methods.

Algorithm and display are not inherently related to any particular computer, virtual system, or other device provided herein. Various general-purpose systems can also be used together with teachings based herein.As described above, it constructs required by this kind of system Structure be obvious.In addition, the present invention is also not directed to any particular programming language.It should be understood that can use various Programming language realizes summary of the invention described herein, and the description done above to language-specific is to disclose this hair Bright preferred forms.

In the instructions provided here, numerous specific details are set forth.It is to be appreciated, however, that implementation of the invention Example can be practiced without these specific details.In some instances, well known method, structure is not been shown in detail And technology, so as not to obscure the understanding of this specification.

Similarly, it should be understood that in order to simplify the disclosure and help to understand one or more of the various inventive aspects, Above in the description of exemplary embodiment of the present invention, each feature of the invention is grouped together into single implementation sometimes In example, figure or descriptions thereof.However, the disclosed method should not be interpreted as reflecting the following intention：It is i.e. required to protect Shield the present invention claims features more more than feature expressly recited in each claim.More precisely, as following Claims reflect as, inventive aspect is all features less than single embodiment disclosed above.Therefore, Thus the claims for following specific embodiment are expressly incorporated in the specific embodiment, wherein each claim itself All as a separate embodiment of the present invention.

Those skilled in the art will understand that can be carried out adaptively to the module in the equipment in embodiment Change and they are arranged in one or more devices different from this embodiment.It can be the module or list in embodiment Member or component are combined into a module or unit or component, and furthermore they can be divided into multiple submodule or subelement or Sub-component.Other than such feature and/or at least some of process or unit exclude each other, it can use any Combination is to all features disclosed in this specification (including adjoint claim, abstract and attached drawing) and so disclosed All process or units of what method or apparatus are combined.Unless expressly stated otherwise, this specification is (including adjoint power Benefit require, abstract and attached drawing) disclosed in each feature can carry out generation with an alternative feature that provides the same, equivalent, or similar purpose It replaces.

In addition, it will be appreciated by those of skill in the art that although some embodiments described herein include other embodiments In included certain features rather than other feature, but the combination of the feature of different embodiments mean it is of the invention Within the scope of and form different embodiments.For example, in the following claims, embodiment claimed is appointed Meaning one of can in any combination mode come using.

Various component embodiments of the invention can be implemented in hardware, or to run on one or more processors Software module realize, or be implemented in a combination thereof.It will be understood by those of skill in the art that can be used in practice In the display device of microprocessor or digital signal processor (DSP) to realize number according to an embodiment of the present invention of breaking one's promise The some or all functions of some or all components.The present invention is also implemented as executing method as described herein Some or all device or device programs (for example, computer program and computer program product).Such reality Existing program of the invention can store on a computer-readable medium, or may be in the form of one or more signals. Such signal can be downloaded from an internet website to obtain, and perhaps be provided on the carrier signal or in any other forms It provides.

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and ability Field technique personnel can be designed alternative embodiment without departing from the scope of the appended claims.In the claims, Any reference symbol between parentheses should not be configured to limitations on claims.Word "comprising" does not exclude the presence of not Element or step listed in the claims.Word "a" or "an" located in front of the element does not exclude the presence of multiple such Element.The present invention can be by means of including the hardware of several different elements and being come by means of properly programmed computer real It is existing.In the unit claims listing several devices, several in these devices can be through the same hardware branch To embody.The use of word first, second, and third does not indicate any sequence.These words can be explained and be run after fame Claim.

The invention also discloses：A1. a kind of display methods for number of breaking one's promise, including：

Word segmentation processing is carried out to the target message got, it is opposite with the target message according to the determination of word segmentation processing result The feature vector answered；

Described eigenvector is inputted into preset SMS classified model, when the output result of the SMS classified model is to lose When believing the result of type, communicating number corresponding with the target message is screened as destination number；

Judge whether the destination number is number of breaking one's promise, if so, for it is described break one's promise number setting in communication into The label of breaking one's promise of row display.

A2. method according to a1, wherein described to judge whether the destination number is specific the step of breaking one's promise number Including：

When monitoring communication event relevant to the destination number, in the corresponding terminal interface of the communication event Show for prompting user break one's promise the prompt information of class label processing for the destination number；

According to the confirmation flag information corresponding with the prompt information received, determine the destination number whether be It breaks one's promise number；

Wherein, the communication event includes：Calling event and/or go electric event.

A3. the method according to A2, wherein the SMS classified model is for determining multiple and different grades of breaking one's promise； When the output result of the SMS classified model be break one's promise type result when, further judge the corresponding mistake of the output result Believe grade；

Then described screen communicating number corresponding with the target message specifically includes for the step of destination number：Root The class information of the destination number is determined according to the corresponding grade of breaking one's promise of the output result；

And it is described for prompt user for the destination number break one's promise class label processing prompt information in further Class information comprising the destination number.

A4. the method according to A2 or A3, wherein the basis receive it is corresponding with the prompt information really Recognize mark information, determines whether the destination number is to specifically include the step of breaking one's promise number：

Determine whether the number of the confirmation flag information corresponding with the destination number received is greater than preset threshold, Determine whether destination number is number of breaking one's promise according to judging result；And/or

Obtain the user of terminal user corresponding to the confirmation flag information corresponding with the destination number received Mark, determines whether the destination number is number of breaking one's promise according to the number of the user identifier；And/or

Temporal information corresponding to the confirmation flag information corresponding with the number of breaking one's promise received is obtained, according to institute The period distribution situation for stating temporal information determines whether the destination number is number of breaking one's promise.

A5. according to any method of A1-A4, wherein it is described for it is described break one's promise number setting in communication into Row display break one's promise label the step of specifically include：

By it is described break one's promise number and it is corresponding with the number of breaking one's promise break one's promise label associated storage into Cloud Server, To be created in communication terminal corresponding with the number of breaking one's promise when monitoring communication event relevant to the number of breaking one's promise Build the display interface for showing the label of breaking one's promise.

A6. method according to a5, wherein described create in communication terminal corresponding with the number of breaking one's promise is used for Break one's promise described in display label display interface the step of after, further comprise：

By break one's promise number and the label of breaking one's promise corresponding with the number of breaking one's promise stored in cloud server It is handed down to the communication terminal corresponding with the number of breaking one's promise, is saved so that the communication terminal is locally downloading, with When just monitoring communication event relevant to the number of breaking one's promise again, break one's promise according to locally saving number and with institute The corresponding label creation of breaking one's promise of number of breaking one's promise is stated for showing the display interface of the label of breaking one's promise.

A7. according to any method of A1-A6, wherein described to judge whether the destination number is number of breaking one's promise Step further comprises：

By the storage of preset law court's class official information into official's database, the mesh is judged in conjunction with official's database Whether label code is number of breaking one's promise.

A8. according to any method of A1-A7, wherein described eigenvector includes：Short message term vector, then it is described right The target message that gets carries out word segmentation processing, according to word segmentation processing result determine feature corresponding with the target message to The step of amount, specifically includes：

Word segmentation processing is carried out to the target message, to obtain at least one participle of the target message；

Each participle input term vector tool at least one described participle is obtained at least one term vector respectively；

By all term vectors and the acquisition of preset corpus word matrix multiple and the mesh at least one described term vector Mark the corresponding short message term vector of short message.

A9. according to any method of A1-A8, wherein described eigenvector includes：Theme vector then obtains for described pair The target message got carries out word segmentation processing, determines feature vector corresponding with the target message according to word segmentation processing result The step of specifically include：

According at least one participle and preset semantic topic for obtain after word segmentation processing to the target message It generates model and obtains theme vector corresponding with the target message.

A10. according to any method of A1-A9, wherein described eigenvector includes：Numeric type vector, then it is described right The target message that gets carries out word segmentation processing, according to word segmentation processing result determine feature corresponding with the target message to The step of amount, specifically includes：

The each participle for including in the target message and digital information are obtained, on the basis of the digital information respectively Calculate the distance that each participle arrives each digital information；

Using the distance as weight, each participle is used as dimension, is described short message content to generate and the mesh Mark the corresponding numeric type vector of short message.

A11. according to any method of A1-A10, wherein the SMS classified model determines in the following manner：

Word segmentation processing is carried out to the sample short message got, it is opposite with the sample short message according to the determination of word segmentation processing result The feature vector answered；

According to feature vector corresponding with the sample short message training SMS classified model；

Wherein, the corresponding feature vector of the sample short message includes：Short message term vector, theme vector, and/or numeric type Vector.

A12. the method according to A11, wherein the sample short message includes：Dun short message, law court or lawyer's letter is short Letter, and/or credit platform short message, the number of breaking one's promise include：Lao Lai number.

B13. a kind of display device for number of breaking one's promise, including：

Word segmentation module, suitable for carrying out word segmentation processing to the target message got, according to the determination of word segmentation processing result and institute State the corresponding feature vector of target message；

Screening module, suitable for described eigenvector is inputted preset SMS classified model, when the SMS classified model Output result be break one's promise type result when, communicating number corresponding with the target message is screened as destination number；

Judgment module, suitable for judging whether the destination number is number of breaking one's promise, if so, being used for the number setting of breaking one's promise In the label of breaking one's promise shown in communication.

B14. device according to b13, wherein the judgment module is particularly adapted to：When monitoring and the target number When the relevant communication event of code, show in the corresponding terminal interface of the communication event for prompting user for the target number Code break one's promise the prompt information of class label processing；According to confirmation flag corresponding with the prompt information letter received Breath, determines whether the destination number is number of breaking one's promise；

Wherein, the communication event includes：Calling event and/or go electric event.

B15. device according to b14, wherein the SMS classified model is for determining multiple and different break one's promise Grade；When the output result of the SMS classified model be break one's promise type result when, the screening module is further adapted for：Judgement The corresponding grade of breaking one's promise of the output result, according to the corresponding grade of breaking one's promise of the output result determine the destination number etc. Grade information；

And it is described for prompt user for the destination number break one's promise class label processing prompt information in further Class information comprising the destination number.

B16. the device according to B14 or B15, wherein the judgment module is particularly adapted to：

Determine whether the number of the confirmation flag information corresponding with the destination number received is greater than preset threshold, Determine whether destination number is number of breaking one's promise according to judging result；And/or

Obtain the user of terminal user corresponding to the confirmation flag information corresponding with the destination number received Mark, determines whether the destination number is number of breaking one's promise according to the number of the user identifier；And/or

Temporal information corresponding to the confirmation flag information corresponding with the number of breaking one's promise received is obtained, according to institute The period distribution situation for stating temporal information determines whether the destination number is number of breaking one's promise.

B17. according to any device of B13-B16, wherein the judgment module is particularly adapted to：

By it is described break one's promise number and it is corresponding with the number of breaking one's promise break one's promise label associated storage into Cloud Server, To be created in communication terminal corresponding with the number of breaking one's promise when monitoring communication event relevant to the number of breaking one's promise Build the display interface for showing the label of breaking one's promise.

B18. the device according to B17, wherein the judgment module is further adapted for：

By break one's promise number and the label of breaking one's promise corresponding with the number of breaking one's promise stored in cloud server It is handed down to the communication terminal corresponding with the number of breaking one's promise, is saved so that the communication terminal is locally downloading, with When just monitoring communication event relevant to the number of breaking one's promise again, break one's promise according to locally saving number and with institute The corresponding label creation of breaking one's promise of number of breaking one's promise is stated for showing the display interface of the label of breaking one's promise.

B19. according to any device of B13-B18, wherein the judgment module is further adapted for：

By the storage of preset law court's class official information into official's database, the mesh is judged in conjunction with official's database Whether label code is number of breaking one's promise.

B20. according to any device of B13-B19, wherein described eigenvector includes：Short message term vector, then it is described Word segmentation module is particularly adapted to：

Word segmentation processing is carried out to the target message, to obtain at least one participle of the target message；

Each participle input term vector tool at least one described participle is obtained at least one term vector respectively；

By all term vectors and the acquisition of preset corpus word matrix multiple and the mesh at least one described term vector Mark the corresponding short message term vector of short message.

B21. according to any device of B13-B20, wherein described eigenvector includes：Theme vector, then described point Word module is particularly adapted to：

According at least one participle and preset semantic topic for obtain after word segmentation processing to the target message It generates model and obtains theme vector corresponding with the target message.

B22. according to any device of B13-B21, wherein described eigenvector includes：Numeric type vector, then it is described Word segmentation module is particularly adapted to：

The each participle for including in the target message and digital information are obtained, on the basis of the digital information respectively Calculate the distance that each participle arrives each digital information；

Using the distance as weight, each participle is used as dimension, is described short message content to generate and the mesh Mark the corresponding numeric type vector of short message.

B23. according to any device of B13-B22, wherein described device further comprises：Training module；And it is described Training module is particularly adapted to：

Word segmentation processing is carried out to the sample short message got, it is opposite with the sample short message according to the determination of word segmentation processing result The feature vector answered；

According to feature vector corresponding with the sample short message training SMS classified model；

Wherein, the corresponding feature vector of the sample short message includes：Short message term vector, theme vector, and/or numeric type Vector.

B24. device according to b13, wherein the sample short message includes：Dun short message, law court or lawyer's letter is short Letter, and/or credit platform short message, the number of breaking one's promise include：Lao Lai number.

C25. a kind of electronic equipment, including：Processor, memory, communication interface and communication bus, the processor, institute It states memory and the communication interface completes mutual communication by the communication bus；

The memory executes the processor such as storing an at least executable instruction, the executable instruction It breaks one's promise described in any one of A1-A12 the corresponding operation of display methods of number.

D26. a kind of computer storage medium is stored with an at least executable instruction in the storage medium, described to hold The corresponding operation of display methods for the number of breaking one's promise that row instruction executes processor as described in any one of A1-A12.

Claims (10)

1. a kind of display methods for number of breaking one's promise, including：

Word segmentation processing is carried out to the target message got, it is corresponding with the target message according to the determination of word segmentation processing result Feature vector；

Described eigenvector is inputted into preset SMS classified model, when the output result of the SMS classified model is class of breaking one's promise When the result of type, communicating number corresponding with the target message is screened as destination number；

Judge whether the destination number is number of breaking one's promise, if so, being the number setting of breaking one's promise for being shown in communication The label of breaking one's promise shown.

2. described to judge whether the destination number is tool the step of breaking one's promise number according to the method described in claim 1, wherein Body includes：

When monitoring communication event relevant to the destination number, show in the corresponding terminal interface of the communication event For prompting user break one's promise the prompt information of class label processing for the destination number；

According to the confirmation flag information corresponding with the prompt information received, determine whether the destination number is to break one's promise Number；

Wherein, the communication event includes：Calling event and/or go electric event.

3. according to the method described in claim 2, wherein, the SMS classified model is for determining multiple and different break one's promise Grade；When the output result of the SMS classified model be break one's promise type result when, further judge that the output result is corresponding Grade of breaking one's promise；

Then described screen communicating number corresponding with the target message specifically includes for the step of destination number：According to institute State the class information that the corresponding grade of breaking one's promise of output result determines the destination number；

And it is described for prompt user for the destination number break one's promise class label processing prompt information in further include The class information of the destination number.

4. according to the method in claim 2 or 3, wherein the basis receives corresponding with the prompt information Confirmation flag information determines whether the destination number is to specifically include the step of breaking one's promise number：

Determine whether the number of the confirmation flag information corresponding with the destination number received is greater than preset threshold, according to Judging result determines whether destination number is number of breaking one's promise；And/or

The user identifier of terminal user corresponding to the confirmation flag information corresponding with the destination number received is obtained, Determine whether the destination number is number of breaking one's promise according to the number of the user identifier；And/or

Obtain temporal information corresponding to the confirmation flag information corresponding with the number of breaking one's promise that receives, according to it is described when Between the period distribution situation of information determine whether the destination number is number of breaking one's promise.

5. method according to claim 1 to 4, wherein it is described for it is described break one's promise number setting in communication into Row display break one's promise label the step of specifically include：

By it is described break one's promise number and it is corresponding with the number of breaking one's promise break one's promise label associated storage into Cloud Server, so as to When monitoring communication event relevant to the number of breaking one's promise, creates and use in communication terminal corresponding with the number of breaking one's promise It breaks one's promise described in the display display interface of label.

6. according to the method described in claim 5, wherein, described create in communication terminal corresponding with the number of breaking one's promise is used Break one's promise described in the display label display interface the step of after, further comprise：

Stored in cloud server described number is broken one's promise and label of breaking one's promise corresponding with the number of breaking one's promise issues for described To the communication terminal corresponding with the number of breaking one's promise, saved so that the communication terminal is locally downloading, so as to again It is secondary when monitoring communication event relevant to the number of breaking one's promise, break one's promise according to locally saving number and with the mistake The corresponding label creation of breaking one's promise of signal code is for showing the display interface of the label of breaking one's promise.

7. -6 any method according to claim 1, wherein described to judge whether the destination number is number of breaking one's promise Step further comprises：

By the storage of preset law court's class official information into official's database, the target number is judged in conjunction with official's database Whether code is number of breaking one's promise.

8. a kind of display device for number of breaking one's promise, including：

Word segmentation module, suitable for carrying out word segmentation processing to the target message got, according to the determination of word segmentation processing result and the mesh Mark the corresponding feature vector of short message；

Screening module, it is defeated when the SMS classified model suitable for described eigenvector is inputted preset SMS classified model Out result be break one's promise type result when, communicating number corresponding with the target message is screened as destination number；

Judgment module, suitable for judging whether the destination number is number of breaking one's promise, if so, for it is described break one's promise number setting for The label of breaking one's promise shown when communication.

9. a kind of electronic equipment, including：Processor, memory, communication interface and communication bus, the processor, the storage Device and the communication interface complete mutual communication by the communication bus；

The memory executes the processor as right is wanted for storing an at least executable instruction, the executable instruction Ask the corresponding operation of display methods for number of breaking one's promise described in any one of 1-7.

10. a kind of computer storage medium, an at least executable instruction, the executable instruction are stored in the storage medium Processor is set to execute the corresponding operation of display methods such as number of any of claims 1-7 of breaking one's promise.