CN103957516A - Junk short message filtering method and engine - Google Patents
Junk short message filtering method and engine Download PDFInfo
- Publication number
- CN103957516A CN103957516A CN201410200231.7A CN201410200231A CN103957516A CN 103957516 A CN103957516 A CN 103957516A CN 201410200231 A CN201410200231 A CN 201410200231A CN 103957516 A CN103957516 A CN 103957516A
- Authority
- CN
- China
- Prior art keywords
- user
- note
- short message
- classification
- engine
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Information Transfer Between Computers (AREA)
Abstract
The invention provides a junk short message filtering method and a junk short message filtering engine. The junk short message filtering method can comprises the steps that the category of a short message is determined through a short message content classifier, whether the category of the short message is a category in which a user is interested or a category disgusting the user is judged according to a user personalized information configuration table, and the short message is filtered according to a judgment result. According to the junk short message filtering method and engine, a personalized short message filtering system based on short message content is obtained, and the accurate rate of junk short message filtering is increased.
Description
Technical field
The present invention relates to moving communicating field, more specifically, relate to filter method and the equipment of refuse messages.
Background technology
In recent years, along with the popularity rate of mobile phone is more and more higher, and the cost of note was more and more lower, and the behavior that utilizes note to realize marketing, even enforcement swindle gets more and more.That user does not wish to receive or be called as refuse messages with a large amount of note that user has nothing to do.According to statistics, the harassing and wrecking that approximately have 35% mobile phone user to be to some extent refuse messages.The refuse messages that each user on average monthly receives is about 8.According in by the end of March, 2013 the telephone subscriber of China Mobile sum reached 11.46 hundred million statistics and carried out conservative estimation, the refuse messages total amount that China mobile phone user receives average every day will exceed 300,000,000.Refuse messages has become a serious social concern.
For fear of the harassing and wrecking of refuse messages, the user of intelligent mobile phone conventionally selects to install filtering junk short messages engine and avoids harassing and wrecking.But there is very large shortcoming in existing filtering short message technology aspect user individual.As everyone knows, same note or a certain class note may be refuse messages for some users, and may belong to valuable information for other user.For example, real estate advertisement SMS may all belong to junk information for most users, but is but Useful Information for just having the user of the related needs such as house-purchase.And for example, whether the advertisement promotion category information such as some educational trainings, Investment & Financing belongs to refuse messages and also depends on user's interest demand at that time.But, mainly rest on SMS sender number aspect for the technical solution of user individual at present, by user, self-defined black and white lists is set particular number is carried out to rigid judgement.Although the scheme of this black and white lists based on sender number can play certain effect, but can not deal with problems from " user is interested in the information of certain content classification at special time period " this angle, thereby also just cannot complete the user individual filtering junk short messages based on short message content.
Therefore, need a kind of filtering short message mechanism of more effective consideration users ' individualized requirement.
Summary of the invention
In order to realize this object, the invention provides a kind of improved method for filtering spam short messages and filtering junk short messages engine, it not only considers number of sender, also considers the individual demand of user to short message content.
The problem of considering in order to solve existing filtering short message mechanism shortcoming user individual, the present invention proposes: by the Text Classification System of introducing based on statistics, note is carried out to classifying content, catch user interest point and/or dislike point by the feedback information in user's active arrangement or use procedure, thereby on the basis of existing filtering short message system, be aided with personalized SMS classified judgement, provide the final result that more meets user's request.
According to an aspect of the present invention, provide a kind of method for filtering spam short messages.The method can comprise: the classification of determining note by short message content grader; According to user personalized information allocation list, judge whether the classification of described note belongs to the classification that user is interested or dislike; And filter described note according to judged result.
In some embodiments of the invention, described short message content grader can be based on support vector machines model.In other embodiment of the present invention, described short message content grader can be based on model-naive Bayesian.According to embodiments of the invention, described short message content grader is to obtain by the training based on numerous samples.
In some embodiments of the invention, described user personalized information allocation list can be by user's active arrangement.In other embodiment of the present invention, described user personalized information allocation list can configure automatically according to user feedback.In preferred embodiments more of the present invention, can both pass through user's active arrangement user personalized information allocation list, and automatically upgrade user personalized information allocation list by collecting user feedback, thereby constantly improve user personalized information allocation list.
In certain embodiments, the mobile phone mainly for Android system according to the method for filtering short message of the embodiment of the present invention.
In some embodiments of the invention, filter whether described note not only belongs to the interested or dislike of the user judged result of classification according to the classification of note, also according to the sender number of described note.
According to a second aspect of the invention, provide a kind of filtering junk short messages engine.This filtering junk short messages engine can comprise: short message content grader, is configured for the classification of determining note; User personalized information collector unit, is configured for and collects user personalized information with maintenance customer's customized information allocation list; Judging unit, is configured for according to user personalized information allocation list, judges whether the classification of described note belongs to the classification that user is interested or dislike; And filter element, be configured for according to judged result and filter described note.
In some preferred embodiments, filtering junk short messages engine according to the present invention is for the mobile phone of Android system.
In certain embodiments, filtering junk short messages engine according to the present invention is arranged on mobile phone as client.
In further embodiments, filtering junk short messages engine according to the present invention is arranged on server.
According to a third aspect of the invention we, provide a kind of mobile phone, it comprises the filtering junk short messages engine according to the embodiment of the present invention.
According to a forth aspect of the invention, provide a kind of server, itself and mobile phone can be coupled communicatedly, and comprise the filtering junk short messages engine according to the embodiment of the present invention.
Filtering junk short messages scheme of the present invention is equivalent in existing filtering junk short messages system, has added a content of text grader based on statistics, according to the setting of user individual or feedback, input note is shown to comprehensive judgement is as final filtration result.Thereby compared with existing filtering short message mechanism, filtering short message mechanism of the present invention provides more excellent performance aspect user individual.
Brief description of the drawings
By below in conjunction with brief description of the drawings the preferred embodiments of the present invention, will make of the present invention above-mentioned and other objects, features and advantages are clearer, wherein:
Fig. 1 schematically shows the schematic diagram of the application scenarios of the mobile communication system 100 that can use therein the embodiment of the present invention;
Fig. 2 schematically shows according to the flow chart of the method for filtering spam short messages of the embodiment of the present invention;
Fig. 3 schematically shows the flow chart of the method for the short message content grader based on SVM model according to the acquisition of the embodiment of the present invention; And
Fig. 4 has schematically shown according to the block diagram of the filtering junk short messages engine of the embodiment of the present invention.
In institute of the present invention drawings attached, same or analogous structure all identifies with same or analogous Reference numeral.
Embodiment
Describe the present invention in detail referring now to accompanying drawing, shown in the drawings of illustrative embodiment of the present invention, to make those skilled in the art can realize the present invention.Being noted that the following drawings and example do not mean that limits the scope of the present invention to single embodiment, on the contrary by exchange and combine described in some or all of different embodiment or shown in element to form other embodiment be also possible.In addition, in the situation that can partially or completely realizing element-specific of the present invention by known tip assemblies, to only describe in these known tip assemblies in order to understand part assembly essential to the invention, and by the detailed description of omitting other parts in these known tip assemblies, to make the present invention more outstanding.Unless pointed out separately herein, otherwise those skilled in the art should understand that: existing although some embodiments of the present invention are described as with software real form, but the present invention is not limited to this, but also can realize with the combination of hardware, software and hardware, and vice versa.Unless explicit state separately herein, otherwise in this manual, it is restrictive the embodiment that shows single component should not being considered as, but the invention is intended to comprise other embodiment including multiple same components, and vice versa.In addition the equivalent current and that develop in the future that, the present invention comprises the known tip assemblies of quoting as signal herein.
As previously mentioned, same note or a certain class note may be refuse messages for some users, and may belong to valuable information for other user.Therefore the strobe utility that, by the service provider of filtering short message mechanism the black and white lists based on number of sender is set can not meet the individual demand of different user.On the other hand, marketing in actual life/promote the note One's name is legion of class, and can constantly there is new marketing note transmit leg, therefore for each number of sender, black and white lists being set by individual subscriber will be loaded down with trivial details, or even unpractical.
In addition, increasing marketing note transmit leg (for example may use providing of buying from operators such as China Mobile, UNICOM, the telecommunications commercial note number that note Batch sending serves, with area code as the number of 021,075 beginning, or with the number of 106 beginnings) send note.Therefore, the mechanism of the black and white lists of the existing sender number of foundation, easily fails to report and uses these commercial note numbers as the refuse messages in the note of transmit leg.If these commercial note numbers are set to blacklist by user, the recommendation activities note wrong report that user may be wished again to the real businessman receiving is refuse messages.
Therefore, in order to improve filtering junk short messages efficiency, reduce and fail to report and report by mistake, the present invention proposes except considering other factors, also to consider the individual demand of user to short message content in filtering junk short messages mechanism.
Fig. 1 shows according to the schematic diagram of the application scenarios of mobile communication system 100 of the present invention.As shown in Figure 1, mobile communication system 100 can comprise mobile phone 120 and server 110.As example, show in the drawings four mobile phone 120-1,120-2,120-3 and 120-4.But should be appreciated that, system 100 can comprise more or less mobile phone.Mobile phone 120 is connected with server 110 by communication network 130.The example of communication network 130 can include but not limited to: the Internet, mobile communications network.
Server 110 is normally by the application vendor maintenance and management that filtering junk short messages engine is provided.The note that application business men can gather user report by server 110 to be to generate note sample set, and note sample set is processed, and to improve the filter effect of filtering junk short messages engine, reduces wrong report and fails to report.Although a server 110 is only shown in figure, should be appreciated that and can have two or more servers 110.Be also to be understood that server 110 can be independent physical entity, also can be distributed on two or more physical entities.
Mobile phone 120 can be can sending and receiving note any mobile phone.Filtering junk short messages engine according to the embodiment of the present invention can be installed on mobile phone 120.Note not only can comprise SMS message in the present invention, also can comprise multimedia message.In the time that user finds that filtering junk short messages engine is failed to report or reported note by mistake, can use its mobile phone 120 to report to application vendor.Should be appreciated that, the present invention is not limited to the concrete communication protocol of each related mobile phone, can include but not limited to 2G, 3G, 4G, 5G wireless communication technology, WCDMA, CDMA2000, TD-SCDMA wireless technology etc.Different mobile phones can adopt identical communication protocol, also can adopt different communication protocol.The present invention is also not limited to the specific operating system of mobile phone, can include but not limited to Android, iOS, Windows Mobile, Symbian, Windows Phone, Blackberry OS etc.Different mobile phones can adopt identical operating system, also can adopt different operating system.Should be noted that filtering short message scheme of the present invention is particularly suitable for the mobile phone of Android system.
Server 110 and mobile phone 120 can communicate by various wireless communication protocols, comprise 2G, 3G, 4G, 5G network, WCDMA, CDMA2000, TD-SCDMA system, WLAN (wireless local area network) (WLAN), etc.
Below with reference to Fig. 2 and Fig. 3, filtering junk short messages mechanism of the present invention is described.
Fig. 2 schematically shows according to the flow chart of the method for filtering spam short messages 200 of the embodiment of the present invention.In the time receiving certain note, filtering junk short messages engine by manner of execution 200 to judge whether this note (this note is also referred to as note to be evaluated) is refuse messages.
In step S210, determine the classification of note by short message content grader.
Short message content grader refers to the disaggregated model based on statistics in machine learning field, such as common naive Bayesian or SVMs (Support Vector Machine is called for short SVM) grader.Below the main and SVM with good classification performance conventional taking text classification field describes as example.Hereinafter also by this short message content grader based on SVM model referred to as svm classifier device.The principle of SVM is mainly in higher dimensional space, to find out the decision-making hyperplane of class interval maximum.
The realization of svm classifier device substantially can be divided into training and differentiate two parts.Before training, need the SMS classified system of model.Complete taxonomic hierarchies carries out the accurately prerequisite of classification by being.This SMS classified system can obtain according to a large amount of note sample statistics.For example,, according to note can be divided into more than ten common classifications such as real estate, educational training, Investment & Financing, tourism trip to the statistics of numerous note samples.
Below, specifically illustrate according to the training process of the SVM short message content grader of the embodiment of the present invention with reference to figure 3.
In step S310, collect the note sample of each classification.Each note sample is endowed class label.Such as real estate class is made as 1, educational training is made as 2, etc.Should be appreciated that, can class label be set in various other modes known or exploitation in the future.
In step S320, note sample is carried out to word segmentation processing one by one.As example, the short message content of a real estate classification is as follows: " room rate news! Military Museum walking ten minutes, the hard-cover little private residence of south, Chang'an street hotel type, three subway zero distances.An account in full 800,000! Lower than general residence price! Surprising preferential only 7 days an of private sale! Condescend to inquire: 58936688 ".This note becomes " three subway zero distance accounts in full 800,000 of the hard-cover little private residence of south, ten minutes Chang'an street of room rate news Military Museum walking hotel type preferentially only condescend to inquire 58936688 in 7 days lower than general residence price private sale is surprising " after word segmentation processing.
In step S330, note sample is carried out to Feature Selection.Note sample is after participle, and all Feature Words (keyword) that obtain have just formed characteristic vector space.Consider because the Spatial Dimension that all Feature Words forms is too high, can only choose those high part dimensions of importance for classification in order to simplify computation complexity.For example, this Feature Selection can realize by following common method: term frequency (Term Frequency, be called for short " TF "), term frequency-contrary text frequency (Term Frequency – Inverse Document Frequency, be called for short " TF-IDF "), the method such as information gain (Information Gain, be called for short " IG ").Preferably, for this short text of note, can choose the simplest TF method.In TF method, choose TF value rank for example, at the word of front N (, choosing N=1000 in experiment) as final characteristic vector.
In step S340, form SVM model.Completed Feature Selection in step S330 after, original short message text just can be converted into the sample point one by one in characteristic vector space.After all sample point inputs, calculate and can find out categorised decision face through SVM model, thereby complete the training of model, obtain SVM model main body.
Alternatively, in order to obtain better classifying quality, can in optional step S350, be optimized model parameter C (slack variable) and gamma (RBF core institute containing parameter).For example, can adopt cross validation (cross validation) through repeatedly training, select C and the gamma with best classification accuracy.
Through above-mentioned steps, can complete the training of SVM model, so method 300 finishes.
This SVM model training can be used as the short message content grader using according to the filtering junk short messages engine of the embodiment of the present invention.For example, in step S210, in the time receiving note, can be by this note input SVM model to be evaluated, through calculating the discriminant classification result of this note.Experiment showed, in real system evaluation and test, according to the SVM model training, note is classified, classification accuracy can reach more than 98%.
More contents about SVM model can be with reference to the paper of Joachims T. " Text categorization with support vector machines:learning with many relevant features ", Proceedings of10th European Conference on Machine Learning (ECML-98), Chemnitz, DE, 1998:137-142.At this, by reference its full content is incorporated to herein.
In step S220, according to user personalized information allocation list, judge whether the classification of note to be evaluated belongs to the classification that user is interested or dislike.
As mentioned above, in different time sections, the interest level to different classes of information or dislike degree are different to different user.How can farthest accurately capture these user individual key elements, be the key that is related to the final performance of filtering short message system.The present invention proposes, and user's customized information allocation list is followed the tracks of user's point of interest or dislike point.According to embodiments of the invention, for example, can catch by following two kinds of modes user's point of interest or dislike point.
(1) allow user initiatively be configured
On application, can allow user's configure user customized information allocation list voluntarily, directly select note classification interested or dislike.
(2) in use point out user feedback
Because most of user's possibility is uncomfortable or be unwilling to carry out active arrangement, can also in use collect feedback information, for example collect about the information of failing to report or reporting by mistake.For example, for the not note of configuration categories of certain user, filtering short message engine may be judged to be refuse messages and be put into dustbin, and user has the operation that it is restored from dustbin; Or filtering short message engine is judged to be normal note by this certain note and does not add interception, and user has the operation of being pulled into dustbin.Finding that in above-mentioned arbitrary situation, system can be sent information, so that whether prompting user is configured to point of interest or dislike point by such note.By collecting this feedback information, can constantly improve user personalized information allocation list.
Should be appreciated that, above-mentioned two kinds of methods can be used separately also and can be combined with.And, according to the user personalized information allocation list of all right allocating default of filtering short message system of the present invention.
In step S230, can filter note according to the judged result in step S220.In one embodiment, in the time that judged result shows that the classification of note is the user's that configures in user personalized information allocation list classification interested (first differentiates result), this note can be differentiated and do not added interception for normal note.In the time that judged result shows that the classification of note is the user's that configures in user personalized information allocation list dislike classification (second differentiates result), this note can be differentiated for refuse messages, thereby this note is put into dustbin.In the time that judged result shows that classification interested that the classification of note neither belongs to the user who configures in user personalized information allocation list does not belong to user's dislike classification yet (the 3rd differentiates result), can filter this note according to default rule, for example, this class note can be considered as to normal note and not add interception.
Preferably, filtering short message engine of the present invention not only considers that whether the classification of note belongs to the classification that user is interested or dislike, and also considers other filter criterias, as the sender number of note.For example, in the case of existing the black and white lists based on sender number, above-mentioned the 3rd note of differentiating result that belongs to further can be filtered, even also the note that belongs to the first court verdict also further can be filtered.
Because the filter method of the black and white lists based on sender number is known, do not repeat them here.
Fig. 4 has schematically shown according to the block diagram of the filtering junk short messages engine 400 of the embodiment of the present invention.Similar with existing filtering junk short messages engine, filtering junk short messages engine according to the present invention can be used as client and is arranged on mobile phone, also can be arranged on server.
As shown in the figure, filtering junk short messages engine 400 can comprise short message content grader 410,
User personalized information collector unit 420, judging unit 430 and filter element 440.
Short message content grader 410 can be configured for the classification of determining note to be evaluated.As mentioned above, short message content grader 410 can be based on support vector machines model, or can be based on model-naive Bayesian.According to embodiments of the invention, short message content grader 410 is trained and is obtained based on numerous samples.
User personalized information collector unit 420 can be configured for maintenance customer's customized information allocation list.This user personalized information allocation list can be by user's active arrangement.In addition, this user personalized information allocation list can configure automatically according to user feedback.In preferred embodiments more of the present invention, can both pass through user's active arrangement user personalized information allocation list, and automatically upgrade user personalized information allocation list by collecting user feedback, thereby constantly improve user personalized information allocation list.Alternatively, the user personalized information allocation list of user personalized information collector unit 420 all right allocating defaults.
Judging unit 430 can be configured for according to user personalized information allocation list, judges whether the classification of note to be evaluated belongs to the classification that user is interested or dislike.Judged result can comprise:
Belong to the user's who configures in user personalized information allocation list classification interested (first differentiates result); Belong to the user's who configures in user personalized information allocation list dislike classification (second differentiates result); And, when the classification interested that neither belongs to user does not belong to user's dislike classification yet (the 3rd differentiates result).
Filter element 440 can be configured for according to above-mentioned judged result and filter described note.For example, in the time there is the first differentiation result, note is differentiated and do not added interception for normal note.In the time there is the second differentiation result, this note is differentiated for refuse messages, thereby this note is put into dustbin.In the time there is the 3rd differentiation result, according to default rule, this note is filtered, for example this note can be considered as to normal note and not add interception.Preferably, filtering short message engine 400 of the present invention not only considers that whether the classification of note belongs to the classification that user is interested or dislike, and also considers other filter criterias, as the sender number of note.For example, the black and white lists in existence based on sender number, filter element 440 can further filter the above-mentioned the 3rd note (even belonging to the note of the first court verdict) of differentiating result.
Short message content grader 410, user personalized information collector unit 420 and judging unit 430 and filter element 440 can be separately implemented at above-mentioned steps S210,220 and 230 operations that complete, and do not repeat them here.
Alternatively, can also comprise other unit according to the filtering junk short messages engine of the embodiment of the present invention, as receiving element, memory cell etc.This receiving element can be configured for note, field feedback of receiving (or test and appraisal) to be filtered etc.This memory cell can be configured for other data that use in note, field feedback, user personalized information allocation list and/or other filter criterias of storing received and filtering junk short messages process or generate.
Above invention has been described in conjunction with the preferred embodiments.It will be understood by those skilled in the art that the method and apparatus illustrating is only exemplary above.Method of the present invention is not limited to the step and the order that illustrate above.Filtering junk short messages engine of the present invention can comprise the parts more more or less than the parts that illustrate.Those skilled in the art can carry out many variations and amendment according to the instruction of illustrated embodiment.
Equipment of the present invention and parts thereof can be by such as very lagre scale integrated circuit (VLSIC) or gate array, realize such as the semiconductor of logic chip, transistor etc. or such as the hardware circuit of the programmable hardware device of field programmable gate array, programmable logic device etc., also can use the software of being carried out by various types of processors to realize, also can be realized by the combination of above-mentioned hardware circuit and software.
The present invention can realize plurality of advantages.Technical scheme of the present invention is on the basis of conventional garbage filtering short message system, introduce the text classifier in machine learning field, by to SMS classified and with the comparison match of user personalized information configuration, realize a kind of personalized filtering short message system based on short message content, thereby further improved the accuracy rate of filtering junk short messages.
Although it should be appreciated by those skilled in the art that by specific embodiment and described the present invention, scope of the present invention is not limited to these specific embodiments.Scope of the present invention is limited by claims and any equivalents thereof.
Claims (10)
1. a method for filtering spam short messages, comprising:
Determine the classification of note by short message content grader;
According to user personalized information allocation list, judge whether the classification of described note belongs to the classification that user is interested or dislike; And
Filter described note according to judged result.
2. method according to claim 1, wherein said short message content grader is based on support vector machines model.
3. method according to claim 1, wherein said user personalized information allocation list is by user's active arrangement.
4. according to the method described in claim 1 or 3, wherein said user personalized information allocation list upgrades automatically according to user feedback.
5. according to the method described in any one in claim 1-3, it is for the mobile phone of Android system.
6. a filtering junk short messages engine, comprising:
Short message content grader, is configured for the classification of determining note;
User personalized information collector unit, is configured for maintenance customer's customized information allocation list;
Judging unit, is configured for according to user personalized information allocation list, judges whether the classification of described note belongs to the classification that user is interested or dislike; And
Filter element, is configured for according to judged result and filters described note.
7. engine according to claim 6, wherein said short message content grader is based on support vector machines model.
8. engine according to claim 6, wherein said user personalized information allocation list is by user's active arrangement.
9. according to the engine described in claim 6 or 8, wherein said user personalized information allocation list also upgrades automatically according to user feedback.
10. a mobile phone, it comprises according to the engine described in any one in claim 6-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410200231.7A CN103957516A (en) | 2014-05-13 | 2014-05-13 | Junk short message filtering method and engine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410200231.7A CN103957516A (en) | 2014-05-13 | 2014-05-13 | Junk short message filtering method and engine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103957516A true CN103957516A (en) | 2014-07-30 |
Family
ID=51334719
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410200231.7A Pending CN103957516A (en) | 2014-05-13 | 2014-05-13 | Junk short message filtering method and engine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103957516A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104484351A (en) * | 2014-11-28 | 2015-04-01 | 上海百事通信息技术股份有限公司 | Large data volume number filtering device and method |
| CN105101124A (en) * | 2015-08-07 | 2015-11-25 | 北京奇虎科技有限公司 | Method and device for marking category of short messages |
| CN105302877A (en) * | 2015-10-09 | 2016-02-03 | 小米科技有限责任公司 | Short message field classification method as well as short message field identification method and apparatus |
| CN105404670A (en) * | 2015-11-16 | 2016-03-16 | 北京奇虎科技有限公司 | Harassing text message determining method and apparatus |
| CN105791543A (en) * | 2016-02-23 | 2016-07-20 | 北京奇虎科技有限公司 | Method, device, client and system for cleaning short messages |
| CN106161731A (en) * | 2015-03-25 | 2016-11-23 | 北京奇虎科技有限公司 | Harassing and wrecking contact person information updating method, terminal unit, server and system |
| CN107517452A (en) * | 2017-09-04 | 2017-12-26 | 上海连尚网络科技有限公司 | A kind of method, equipment and computer-readable storage medium for being used to manage short message |
| CN107786736A (en) * | 2017-10-16 | 2018-03-09 | 微梦创科网络科技（中国）有限公司 | A kind of intelligent control method and control system of refuse messages alerting pattern |
| CN108804669A (en) * | 2018-06-08 | 2018-11-13 | 国家计算机网络与信息安全管理中心 | A kind of fraudulent call method for detecting based on intention understanding technology |
| CN108833720A (en) * | 2018-05-04 | 2018-11-16 | 北京邮电大学 | Fraudulent call number identification method and system |
| CN110972086A (en) * | 2018-09-29 | 2020-04-07 | 北京奇虎科技有限公司 | Short message processing method and device, electronic equipment and computer readable storage medium |
| CN116074768A (en) * | 2022-12-30 | 2023-05-05 | 南京叁月半信息科技有限公司 | Communication method, device and equipment of mobile communication terminal |
| CN117278948A (en) * | 2023-09-18 | 2023-12-22 | 江苏力争信息科技有限公司 | System and method for discriminating short message type of enterprise short message |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101257671A (en) * | 2007-07-06 | 2008-09-03 | 浙江大学 | Method for real time filtering large scale rubbish SMS based on content |
| CN102355517A (en) * | 2011-07-01 | 2012-02-15 | 宇龙计算机通信科技(深圳)有限公司 | Information classification apparatus, information classification method and terminal |
| US8214437B1 (en) * | 2003-07-21 | 2012-07-03 | Aol Inc. | Online adaptive filtering of messages |
| CN103037339A (en) * | 2012-12-28 | 2013-04-10 | 深圳市彩讯科技有限公司 | Short message filtering method based on user creditworthiness and short message spam degree |
| CN103778225A (en) * | 2014-01-23 | 2014-05-07 | 北京奇虎科技有限公司 | Processing method, identifying device and identifying system of advertisement marketing language information |
-
2014
- 2014-05-13 CN CN201410200231.7A patent/CN103957516A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8214437B1 (en) * | 2003-07-21 | 2012-07-03 | Aol Inc. | Online adaptive filtering of messages |
| CN101257671A (en) * | 2007-07-06 | 2008-09-03 | 浙江大学 | Method for real time filtering large scale rubbish SMS based on content |
| CN102355517A (en) * | 2011-07-01 | 2012-02-15 | 宇龙计算机通信科技(深圳)有限公司 | Information classification apparatus, information classification method and terminal |
| CN103037339A (en) * | 2012-12-28 | 2013-04-10 | 深圳市彩讯科技有限公司 | Short message filtering method based on user creditworthiness and short message spam degree |
| CN103778225A (en) * | 2014-01-23 | 2014-05-07 | 北京奇虎科技有限公司 | Processing method, identifying device and identifying system of advertisement marketing language information |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104484351B (en) * | 2014-11-28 | 2018-07-20 | 上海百事通信息技术股份有限公司 | Big data quantity number filtering device and method |
| CN104484351A (en) * | 2014-11-28 | 2015-04-01 | 上海百事通信息技术股份有限公司 | Large data volume number filtering device and method |
| CN106161731A (en) * | 2015-03-25 | 2016-11-23 | 北京奇虎科技有限公司 | Harassing and wrecking contact person information updating method, terminal unit, server and system |
| CN105101124A (en) * | 2015-08-07 | 2015-11-25 | 北京奇虎科技有限公司 | Method and device for marking category of short messages |
| CN105302877A (en) * | 2015-10-09 | 2016-02-03 | 小米科技有限责任公司 | Short message field classification method as well as short message field identification method and apparatus |
| CN105302877B (en) * | 2015-10-09 | 2019-03-08 | 小米科技有限责任公司 | For the method for short message domain classification, the recognition methods of short message field and device |
| CN105404670A (en) * | 2015-11-16 | 2016-03-16 | 北京奇虎科技有限公司 | Harassing text message determining method and apparatus |
| CN105404670B (en) * | 2015-11-16 | 2018-09-25 | 北京奇虎科技有限公司 | Harass short message method of discrimination and device |
| CN105791543A (en) * | 2016-02-23 | 2016-07-20 | 北京奇虎科技有限公司 | Method, device, client and system for cleaning short messages |
| WO2019042164A1 (en) * | 2017-09-04 | 2019-03-07 | 上海连尚网络科技有限公司 | Method, apparatus, and computer storage medium for managing sms messages |
| CN107517452A (en) * | 2017-09-04 | 2017-12-26 | 上海连尚网络科技有限公司 | A kind of method, equipment and computer-readable storage medium for being used to manage short message |
| CN107786736A (en) * | 2017-10-16 | 2018-03-09 | 微梦创科网络科技（中国）有限公司 | A kind of intelligent control method and control system of refuse messages alerting pattern |
| CN108833720A (en) * | 2018-05-04 | 2018-11-16 | 北京邮电大学 | Fraudulent call number identification method and system |
| CN108804669A (en) * | 2018-06-08 | 2018-11-13 | 国家计算机网络与信息安全管理中心 | A kind of fraudulent call method for detecting based on intention understanding technology |
| CN110972086A (en) * | 2018-09-29 | 2020-04-07 | 北京奇虎科技有限公司 | Short message processing method and device, electronic equipment and computer readable storage medium |
| CN116074768A (en) * | 2022-12-30 | 2023-05-05 | 南京叁月半信息科技有限公司 | Communication method, device and equipment of mobile communication terminal |
| CN116074768B (en) * | 2022-12-30 | 2024-05-14 | 中贸区块链(深圳)有限公司 | Communication method, device and equipment of mobile communication terminal |
| CN117278948A (en) * | 2023-09-18 | 2023-12-22 | 江苏力争信息科技有限公司 | System and method for discriminating short message type of enterprise short message |
| CN117278948B (en) * | 2023-09-18 | 2024-05-14 | 江苏力争信息科技有限公司 | System and method for discriminating short message type of enterprise short message |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103957516A (en) | Junk short message filtering method and engine | |
| CN103607705B (en) | Method for filtering spam short messages and engine | |
| CN102415080B (en) | Method and system for targeted offers to mobile users | |
| US20100145771A1 (en) | System and method for providing service or adding benefit to social networks | |
| CN106354797B (en) | Data recommendation method and device | |
| CN109978608A (en) | The marketing label analysis extracting method and system of target user's portrait | |
| US20070233861A1 (en) | Method and apparatus for implementing SMS SPAM filtering | |
| US8229415B1 (en) | Wireless communication data store construction and analysis | |
| CN105007171A (en) | User data analysis system and method based on big data in communication field | |
| CN102801817B (en) | Subscriber context-based pushing method and device | |
| CN105469502A (en) | Intelligent oiling method and terminal | |
| EP2756432A1 (en) | System and method for automated classification of web pages and domains | |
| CN105630977A (en) | Application recommending method, device and system | |
| CN110046201B (en) | Method, device and system for processing general ledger subject data of business transaction | |
| TW201250611A (en) | Message delivery system with consumer attributes collecting mechanism and transaction history recording mechanism and communication system using same | |
| WO2010131259A2 (en) | A monitoring system for capturing information relating to the wireless smart device for behavioral analytics | |
| CN103634470A (en) | Human-computer interaction prediction method based on terminal mobile data access network Qos | |
| CN104182470B (en) | A kind of mobile terminal application class system and method based on SVM | |
| EP2493150B1 (en) | Mobile application metadata collection agent | |
| CN105554785A (en) | Grouping method and grouping device | |
| CN109391913A (en) | A kind of method and system based on the slice management of NB-IoT Internet resources | |
| CN103107936B (en) | A kind of method and device sending information | |
| CN107257312B (en) | A kind of data processing method and device | |
| CN110445879A (en) | The control method and device of pushed information are unified on a kind of backstage | |
| CN104254105A (en) | Forward load control method and equipment for mobile Internet business |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20140730 |
|
| WD01 | Invention patent application deemed withdrawn after publication |