CN102111767A - Method for improving correct rate of identifying junk short message number based on called dispersed degree - Google Patents
Method for improving correct rate of identifying junk short message number based on called dispersed degree Download PDFInfo
- Publication number
- CN102111767A CN102111767A CN2009102006514A CN200910200651A CN102111767A CN 102111767 A CN102111767 A CN 102111767A CN 2009102006514 A CN2009102006514 A CN 2009102006514A CN 200910200651 A CN200910200651 A CN 200910200651A CN 102111767 A CN102111767 A CN 102111767A
- Authority
- CN
- China
- Prior art keywords
- note
- called
- threshold value
- refuse messages
- calling number
- 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
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a method for improving correct rate of identifying a junk short message number based on a called dispersed degree. The method comprises the following steps: setting a called dispersed degree threshold value; in a set time range, recording each short message sent by a calling number, and increasing a called number in attributes of each message; when the number of the messages sent by the calling number in the set time range exceeds the threshold value, calculating the called dispersed degree of the message sent by the calling number, namely the percentage of called numbers and total message numbers in the sent short messages; when the called dispersed degree exceeds the set called dispersed degree threshold value, regarding the calling number as the junk short message number; and when the called dispersed degree does not exceed the set called dispersed degree threshold value, considering that the calling number is not the junk short message number. By using the method, when short message flow rate is in over frequency, the sending behavior of a subscriber is analyzed according to the characteristics of the called number, thereby expelling normal short message numbers, reducing interception-failure efficiency of intercepting the junk short messages and improving the correct rate of indentifying the junk short messages.
Description
Technical field
The present invention relates to a kind of method that improves refuse messages number recognition accuracy when realizing, relate in particular to a kind of method of the raising refuse messages number recognition correct rate based on called dispersion by the traffic statistics note.
Background technology
SMS (Short Message Service) is as a kind of basic service of mobile communications network, and when convenient message communicating service was provided for the user, also the propagation for garbage information provided channel.And rubbish short message has the trend that grows in intensity, and refuse messages not only brings the harmful effect of customer complaint, also has the malicious owing fee problem, therefore need monitor interception in real time to refuse messages.
The time frequency traffic characteristic of refuse messages comprises: 1, send note time overlength continuously; 2, the unit interval amount greatly; 3, called number disperses.
At present, according to the temporal characteristics of refuse messages, adopt the method for the note flow that sends by time scope statistics number, identify the note time to send a large amount of short message number, and list the refuse messages number in, limiting this number, to send note be a kind of effective means.
Said method can in time be found the refuse messages number from the temporal characteristics angle of refuse messages.But frequently send the situation of note for normal users, because the frequency ratio that note sends is higher, also caused the normal users number to be put into the refuse messages number, it is unsuccessful to cause the normal users note to send.
Original note flow rate calculation mode is the frequency that the calling number in section computing time sends note, and the excess flow threshold value is the refuse messages number.
See also Fig. 1, the statistics formation schematic diagram of the no called number of prior art, setting flow threshold is 10/1 minute, exceeds this value and is the refuse messages number, number 8613988888888 has reached 10 in 45 seconds of 08:00:00-08:00:45.Number 8613988888888 is listed in the refuse messages number.
Summary of the invention
The objective of the invention is to overcome the defective of prior art and a kind of method of the raising refuse messages number recognition correct rate based on called dispersion is provided, can be when note flow overclocking according to called characteristic analysis user's transmission behavior, get rid of normal note number, reduce intercepting rubbish short message generation elam error rate, improve the refuse messages recognition accuracy.
The technical scheme that realizes above-mentioned purpose is: a kind of method of the raising refuse messages number recognition correct rate based on called dispersion,
Set called dispersion threshold value;
In the time range of setting, every note that the record calling number sends increases called number in every note attribute;
When the note quantity that sends when calling number in the time range of setting exceeds threshold value, calculate the called dispersion that this calling number sends note, promptly send the unduplicated called number number in the note and the percentage of note sum;
When called dispersion exceeds the called dispersion threshold value of setting, think that this calling number is the refuse messages number;
When called dispersion does not reach the called dispersion threshold value of setting, do not think that then this calling number is the refuse messages number.
The method of above-mentioned raising refuse messages number recognition correct rate based on called dispersion, wherein, it may further comprise the steps:
At first, write down time, called number that each calling number sends note, check that whether each bar note that calling number sends has reached the interior flow threshold of fixed time section, is correspondingly processed according to different states;
If be no more than flow threshold, then remove the expired note of this calling number, keep the note in the fixed time section, write down the time and the called number of every note, continue monitoring;
If excess flow threshold value but do not reach the called dispersion of setting is then removed the expired note of this calling number, keep the note in the fixed time section, write down the time and the called number of every note, this number is as the refuse messages number, and continues to monitor;
If excess flow threshold value and exceed the called dispersion of setting is then removed the expired note of this calling number, keep the note in the fixed time section, write down the time and the called number of every note, list in the refuse messages number as the refuse messages number.
The invention has the beneficial effects as follows: the present invention is a kind of method that improves refuse messages number accuracy rate during by the traffic statistics note, merely according to the method difference of traffic statistics, carries out secondary calculating with tradition after the overclocking.Adopted this method, during the flow overclocking, added up called dispersion once more, can filter point-to-point normal note, avoided wrong refuse messages number that right number is listed in.
Description of drawings
Fig. 1 is the statistics formation schematic diagram of the no called number of prior art;
Fig. 2 is the flow chart of one embodiment of the invention;
Fig. 3 is the statistics formation schematic diagram that called number is arranged of embodiments of the invention.
Embodiment
A kind of method of the raising refuse messages number recognition correct rate based on called dispersion, implementation method is: set called dispersion threshold value; In the time range of setting, every note that the record calling number sends increases called number in every note attribute; When the note quantity that sends when calling number in the time range of setting exceeds threshold value, calculate the called dispersion that this calling number sends note, promptly send the unduplicated called number number in the note and the percentage of note sum; When called dispersion exceeds the called dispersion threshold value of setting, think that this calling number is the refuse messages number; When called dispersion does not reach the called dispersion threshold value of setting, do not think that then this calling number is the refuse messages number.
This method may further comprise the steps:
At first, write down time, called number that each calling number sends note, check that whether each bar note that calling number sends has reached the interior flow threshold of fixed time section, is correspondingly processed according to different states;
If be no more than flow threshold, then remove the expired note of this calling number, keep the note in the fixed time section, write down the time and the called number of every note, continue monitoring;
If excess flow threshold value but do not reach the called dispersion of setting is then removed the expired note of this calling number, keep the note in the fixed time section, write down the time and the called number of every note, this number is as the refuse messages number, and continues to monitor;
If excess flow threshold value and exceed the called dispersion of setting is then removed the expired note of this calling number, keep the note in the fixed time section, write down the time and the called number of every note, list in the refuse messages number as the refuse messages number.
The invention will be further described below in conjunction with an embodiment.
See also Fig. 2, the flow chart for one embodiment of the invention may further comprise the steps:
Step S1 receives note;
Step S2 deposits the note formation by calling number in;
Step S3 judges whether excess flow threshold value of this calling number,
If then enter step S4;
If not, then return step S1;
Step S4 judges whether the called dispersion of this calling number exceeds threshold value,
If then enter step S5;
If not, then return step S1;
Step S5 confirms, promptly confirms as the refuse messages number.
See also Fig. 3, the statistics formation schematic diagram that called number is arranged of the present invention, the present invention has increased earlier the called number field on original statistical, calculate the dispersion of called number; And set the suitable numerical value (as 80%) of dispersion, exceed this numerical value and be judged to be the refuse messages number.
In the note that calling number sends among Fig. 3 not the repeated number yardage be 1; Only comprise 8,613,911,111,111 1 numbers, the note number is 10, the number dispersion: repeated number yardage/note number=1/10=10% not.
So, though this calling number excess flow threshold value, not as the refuse messages number.Adopt the accuracy that called dispersion is filtered has increased the refuse messages number.
In sum, after the present invention sends the excess flow threshold value to note in the time period, and then add up for called subscriber's distribution situation, according to calculating the percentage that the user sends the called number number/transmission note sum of note in the monitoring period, to exceeding the user who sets percentage, list the refuse messages user in., when number sends overclocking, and then calculate called dispersion and improve accuracy rate during the period in monitoring.
Specifically, when the note number that sends in the official hour scope when number exceeds prior preset threshold, by calculating the dispersion (not repeated number yardage/note number) of called number in the note that sends, filter out normal note number, identify the refuse messages number, improve refuse messages number recognition correct rate.
The present invention can be applicable to the note optimization system, can significantly reduce the elam error rate of refuse messages, improves the degree of hitting of refuse messages identification.
Below embodiment has been described in detail the present invention in conjunction with the accompanying drawings, and those skilled in the art can make the many variations example to the present invention according to the above description.Thereby some details among the embodiment should not constitute limitation of the invention, and the scope that the present invention will define with appended claims is as protection scope of the present invention.
Claims (2)
1. the method based on the raising refuse messages number recognition correct rate of called dispersion is characterized in that,
Set called dispersion threshold value;
In the time range of setting, every note that the record calling number sends increases called number in every note attribute;
When the note quantity that sends when calling number in the time range of setting exceeds threshold value, calculate the called dispersion that this calling number sends note, promptly send the unduplicated called number number in the note and the percentage of note sum;
When called dispersion exceeds the called dispersion threshold value of setting, think that this calling number is the refuse messages number;
When called dispersion does not reach the called dispersion threshold value of setting, do not think that then this calling number is the refuse messages number.
2. the method for the raising refuse messages number recognition correct rate based on called dispersion according to claim 1 is characterized in that it may further comprise the steps:
At first, write down time, called number that each calling number sends note, check that whether each bar note that calling number sends has reached the interior flow threshold of fixed time section, is correspondingly processed according to different states;
If be no more than flow threshold, then remove the expired note of this calling number, keep the note in the fixed time section, write down the time and the called number of every note, continue monitoring;
If excess flow threshold value but do not reach the called dispersion of setting is then removed the expired note of this calling number, keep the note in the fixed time section, write down the time and the called number of every note, this number is as the refuse messages number, and continues to monitor;
If excess flow threshold value and exceed the called dispersion of setting is then removed the expired note of this calling number, keep the note in the fixed time section, write down the time and the called number of every note, list in the refuse messages number as the refuse messages number.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009102006514A CN102111767A (en) | 2009-12-24 | 2009-12-24 | Method for improving correct rate of identifying junk short message number based on called dispersed degree |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009102006514A CN102111767A (en) | 2009-12-24 | 2009-12-24 | Method for improving correct rate of identifying junk short message number based on called dispersed degree |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102111767A true CN102111767A (en) | 2011-06-29 |
Family
ID=44175763
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009102006514A Pending CN102111767A (en) | 2009-12-24 | 2009-12-24 | Method for improving correct rate of identifying junk short message number based on called dispersed degree |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102111767A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103139730A (en) * | 2011-11-23 | 2013-06-05 | 上海粱江通信系统股份有限公司 | Method used for identifying situation of mass numbers sending junk short messages at low frequency |
| CN103167501A (en) * | 2011-12-15 | 2013-06-19 | 上海粱江通信系统股份有限公司 | Method for improving identification accuracy rates of crank call number based on called dispersion degree |
| CN109104702A (en) * | 2017-06-20 | 2018-12-28 | 中兴通讯股份有限公司 | Information intercepting method, device and storage medium |
-
2009
- 2009-12-24 CN CN2009102006514A patent/CN102111767A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103139730A (en) * | 2011-11-23 | 2013-06-05 | 上海粱江通信系统股份有限公司 | Method used for identifying situation of mass numbers sending junk short messages at low frequency |
| CN103139730B (en) * | 2011-11-23 | 2016-03-30 | 上海粱江通信系统股份有限公司 | For identifying that a large amount of number low frequency sends the method for refuse messages situation |
| CN103167501A (en) * | 2011-12-15 | 2013-06-19 | 上海粱江通信系统股份有限公司 | Method for improving identification accuracy rates of crank call number based on called dispersion degree |
| CN109104702A (en) * | 2017-06-20 | 2018-12-28 | 中兴通讯股份有限公司 | Information intercepting method, device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101790142B (en) | Method and system for identifying spam message sources by combining message contents and transmission frequency | |
| CN101771966B (en) | Keywords and frequency based method for identifying spam message sources | |
| CN102111731A (en) | Method based on content similarity for improving recognition accuracy of spam message numbers | |
| WO2016065908A1 (en) | Method, device and system for detecting fraudulent user | |
| WO2010031294A1 (en) | De-massing method of position advertising service based on regional strategy and system thereof | |
| CN101217820A (en) | An identification system and identification method on disturbance telephone numbers | |
| CN101150762A (en) | A spam real time interception method and system | |
| CN102761872A (en) | Spam message intercepting method | |
| CN101321070B (en) | Monitoring system and method for suspicious user | |
| WO2010054564A1 (en) | Flow control method and corresponding system for cell short message | |
| CN101909261A (en) | Method and system for monitoring spam | |
| CN101472247A (en) | Method and system for controlling rubbish short message | |
| CN102231888A (en) | Monitoring method and device | |
| CN102111767A (en) | Method for improving correct rate of identifying junk short message number based on called dispersed degree | |
| CN102111723B (en) | Method for identifying spam short message user by analyzing short message frequency and content | |
| CN110072251B (en) | Method and device for analyzing user communication behavior and managing user | |
| CN103733581B (en) | Message processing method and base station | |
| WO2011140874A1 (en) | Method and apparatus for evaluating behavior of user equipment in standby state | |
| CN103139730B (en) | For identifying that a large amount of number low frequency sends the method for refuse messages situation | |
| CN101827328A (en) | Device and method for monitoring short-message | |
| CN108259363B (en) | Method and device for controlling stepped service flow | |
| CN102572746B (en) | A kind of method sending behavioural characteristic identification junk short message source based on the frequency and user | |
| CN103167501A (en) | Method for improving identification accuracy rates of crank call number based on called dispersion degree | |
| CN104168547A (en) | A system and method for processing fraud short messages based on signaling technology | |
| CN101340693B (en) | System and implementing method for monitoring rubbish short message based on content length |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20110629 |