CN102193948A - Feature matching method and device - Google Patents
Feature matching method and device Download PDFInfo
- Publication number
- CN102193948A CN102193948A CN2010101271638A CN201010127163A CN102193948A CN 102193948 A CN102193948 A CN 102193948A CN 2010101271638 A CN2010101271638 A CN 2010101271638A CN 201010127163 A CN201010127163 A CN 201010127163A CN 102193948 A CN102193948 A CN 102193948A
- Authority
- CN
- China
- Prior art keywords
- matching
- feature data
- history feature
- sent
- input data
- 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
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention relates to a feature matching method and device. The feature matching method comprises the following steps of: performing feature matching on received input data according to a learnt history feature data set; if history feature data fully matched with the input data exists in the history feature data set, acquiring a matching result according to the history feature data matched with the input data; and if no history feature data fully matched with the input data exists in the history feature data set, wholly or partially sending the input data to a matching engine to perform feature matching. According to the method provided by the embodiment of the invention, the input data is compared with the learnt history feature data set, and the matching result is directly acquired if the history feature data fully matched with the input data exists in the history feature data set, thereby reducing the frequency of accessing a feature library and increasing the matching speed and efficiency.
Description
Technical field
The embodiment of the invention relates to communication technical field, particularly a kind of feature matching method and device.
Background technology
Along with rapid development of Internet, the class of business of network is more and more, and new business emerges in an endless stream and becomes increasingly complex, and new problem is also more and more.For example: aspect network control and Bandwidth Management, the P2P business has occupied 70% network bandwidth resources, have or even the malice of the network bandwidth taken, had a strong impact on user's experience; Aspect network security, network intrusions and attack and more and more to concentrate on application layer, common fire wall for the virus disseminating among IP bag payload of hiding oneself, attack function a little less than; In addition, operator also has the demand by content charging for fear of the embarrassment of becoming " pipeline merchant ".
In order to address these problems, need to discern the packet application layer content on the network, thereby produce deep-packet detection (Deep Packet Inspection; Be called for short: DPI) technology.The DPI technology is: the application layer content to different business is analyzed, and extracts certain professional feature rule of the unique difference of energy; Packet on the network and this feature rule are mated,, then can identify this business, and then carry out the operation corresponding with this business if the match is successful.In the existing DPI technology, when matching engine receives network packet, a part (or all) data of network packet and the data structure in the feature database are mated.The data volume of feature database is big, generally is stored in outside the sheet.
The inventor finds prior art in realizing process of the present invention there are the following problems at least:
When matching engine received the network packet of repetition, matching operation repeated, and the acting frequently of access characteristic storehouse causes that matching speed is slow, efficient is low.
Summary of the invention
The embodiment of the invention provides a kind of feature matching method and device, in order to solve existing slow, the inefficient problem of characteristic matching speed, improves matching speed and efficient.
The embodiment of the invention provides a kind of feature matching method, comprising:
According to the history feature data acquisition of having learnt, the input data that receive are carried out characteristic matching;
If there are the history feature data of all mating in the described history feature data acquisition, then according to obtaining matching result with the history feature data of described input Data Matching with described input data;
If do not have the history feature data of all mating in the described history feature data acquisition, then all or part of matching engine that is sent to of described input data carried out characteristic matching with described input data.
The embodiment of the invention provides a kind of characteristic matching device again, comprising: memory letter sorting unit and matching engine; Described memory letter sorting unit comprises: data set matching module and judging module;
Described data set matching module is used for according to the history feature data acquisition of having learnt the input data that receive being carried out characteristic matching;
Described judging module is used for if there are the history feature data of all mating with described input data in described history feature data acquisition, then according to obtaining matching result with the history feature data of described input Data Matching; If do not exist in the described history feature data acquisition and the whole history feature data of coupling of described input data, then described input data all or part of is sent to described matching engine and carries out characteristic matching;
Described matching engine is used for according to feature database the input data that receive being carried out characteristic matching.
Feature matching method that the embodiment of the invention provides and device, input data and the history feature data acquisition of having learnt are compared, if there are all history feature data of coupling in the history feature data acquisition, can directly obtain matching result, reduce the frequency in access characteristic storehouse, improved matching speed and efficient.
Description of drawings
Fig. 1 is the schematic flow sheet of feature matching method first embodiment of the present invention;
Fig. 2 a is the schematic flow sheet of feature matching method second embodiment of the present invention;
Fig. 2 b is the synoptic diagram of the application scenarios of feature matching method second embodiment of the present invention;
Fig. 3 a is the schematic flow sheet of feature matching method the 3rd embodiment of the present invention;
Fig. 3 b is the synoptic diagram of the application scenarios of feature matching method the 3rd embodiment of the present invention;
Fig. 4 is the structural representation of characteristic matching device first embodiment of the present invention;
Fig. 5 is the synoptic diagram of characteristic matching device first embodiment of the present invention.
Embodiment
Below by drawings and Examples, technical scheme of the present invention is described in further detail.
Fig. 1 is the schematic flow sheet of feature matching method first embodiment of the present invention, and as shown in Figure 1, this feature matching method comprises:
The history feature data acquisition that step 101, basis have been learnt carries out characteristic matching to the input data that receive;
Before execution in step 101, can receive and learn history feature data and corresponding matching result thereof that described matching engine sends, particularly: after matching engine is carried out characteristic matching according to rule in the feature database to the input data, if the match is successful, can with this input data correspondence with feature database in the characteristic that is complementary of rule and matching result send to memory letter sorting unit.After memory letter sorting unit receives the characteristic and matching result that the match is successful, this characteristic can be saved as the history feature data, and preserve the matching result of this history feature data correspondence.Wherein, when memory letter sorting unit is saved in the history feature data in the history feature data acquisition, can adopt certain algorithm, for example: methods such as Hash (hash) algorithm or direct mapping are preserved, then can be: if set up the history feature data acquisition, when the hash conflict occurring with the hash algorithm according to certain regular regular update history feature data acquisition, for example, can replace old hash list item with new hash list item, finish the renewal of history feature data acquisition; Perhaps also can adopt other rule, for example: least frequent service regeulations etc. recently, upgrade the history feature data acquisition.
In the follow-up processing procedure, after memory letter sorting unit received the input data, the history feature data that are complementary with rule in previous that learn and the feature database compared.Wherein, feature database is the feature set of representing with certain data structure, comprises the lot of data structure.Matching engine can comprise the sub-matching engine of one or more serial or parallels.When matching engine receives input during data, part or all of taking out the input data one by one mated with the data structure in the feature database, when the data structure coupling of input data and certain feature of expression upward the time, can export the feature that matches.If the input data are not mated any feature at feature database, then can export not match information.
If certain bar history feature data identical (being that both all mate) in input data and the history feature data acquisition, then memory letter sorting unit can be directly with the matching result of this history feature data correspondence matching result as present input data, present input data does not need to give matching engine again and mates, can directly this matching result be sent to the results management module and handle, determine final output result.For example: if the input data are URL(uniform resource locator) (URL), the existing URL of certain bar in this URL and the history feature data acquisition all mates, then can be with the matching result of this existing URL correspondence matching result as the URL of current input; If the input data are character string, the existing character string of certain bar in this character string and the history feature data acquisition is all mated, then can be with the matching result of this existing character string correspondence matching result as the character string of current input.
Wherein, do not exist the history feature data of all mating with the input data specifically can comprise in the history feature data acquisition: any history feature data in the history feature data acquisition all do not match with the input data, at this moment, can directly these input data be sent to matching engine and carry out characteristic matching; Though perhaps there are not history feature data in the history feature data acquisition with the whole couplings of input data, but have and the history feature data of importing the data division coupling, the unit of memory letter sorting at this moment can be sent to matching engine with the part of not mating in the described input data to carry out characteristic matching and notifies matching engine to begin the position of mating from feature database; Matching engine can be mated the input data part that the match is successful from the position that feature database begins to mate according to this, need not whole input data are mated, thereby can improve matching speed.Wherein, matching engine can be made up of the sub-engine of the coupling of one or more serial or parallels, and each mates sub-engine access characteristic storehouse, determines according to specific algorithm whether the part of needs coupling in the input data can mate the rule in the feature database.If on the coupling, then one side will be imported section data or all reach matching result and return to memory letter sorting unit, record in the history feature data acquisition, on the one hand matching result be sent to the results management module.The results management module then can adopt specific rule to handle according to the memory letter sorting unit that receives and the matching result of matching engine, determines final output result.
Further, step 103 can comprise following example:
Example one, described input data are URL(uniform resource locator) (URL), and described URL(uniform resource locator) comprises host name (host) and path (path).
If history feature data in the described history feature data acquisition and described host name and path all do not match, then described host name and path are sent to matching engine and carry out characteristic matching; Perhaps,
If have the history feature data of all mating in the described history feature data acquisition, described path be sent to matching engine carry out characteristic matching with described host name; Perhaps exist when be the history feature data of part coupling, the part of not mating in described path is sent to matching engine carries out characteristic matching with described path; Perhaps host name and described path are sent to matching engine and carry out characteristic matching.
Wherein, because the host name of the URL(uniform resource locator) of same data stream may be identical, but the identical probability in path is lower, therefore can only preserve the host name that the match is successful in the history feature data acquisition, host name relatively only in the time of relatively sends to matching engine with the path of the URL(uniform resource locator) of host name coupling then and carries out characteristic matching.In addition, also can both preserve the host name that the match is successful, and preserve the path that the match is successful again, at this moment, then can both compare host name, and compare the path again, the part of will not mate is sent to matching engine and carries out characteristic matching then.
Example two, input data are character string.
If history feature data and described character string in the described history feature data acquisition all do not match, then described character string is sent to matching engine and carries out characteristic matching; Perhaps
If do not have the history feature data of all mating in the described history feature data acquisition, but have the history feature data of partly mating, then described character string is sent to matching engine and carries out characteristic matching with described character string with described character string; Perhaps the part that described character string is not mated is sent to matching engine and carries out characteristic matching.
Present embodiment will be imported data and compare with the history feature data acquisition of having learnt, if there are all history feature data of coupling in the history feature data acquisition, can directly obtain matching result, do not need all will import data sends to matching engine and mates at every turn, therefore need not the frequent visit feature database, reduce the frequency in access characteristic storehouse, improved matching speed and efficient.
Fig. 2 a is the schematic flow sheet of feature matching method second embodiment of the present invention, shown in Fig. 2 a, on the basis of feature matching method first embodiment of the present invention, be URL(uniform resource locator) (Uniform/Universal Resource Locator with the input data; Be called for short: URL) be example, URL generally is made up of host name (host) and path (path) two parts, and this feature matching method specifically can comprise:
Wherein, the history feature data acquisition can not set up different subclass according to different flow points, the host of various flows can be set up the data set of a mixing yet.When the item number of the host that writes down in the history feature data acquisition reaches maximal value, can also be according to certain algorithm with old record deletion, as aging by the time, the most seldom using priciple etc. is deleted recently.Table 1 is a kind of form of the host of the URL of history feature data acquisition preservation.
Table 1
| host | Path is in the starting position of feature database coupling | Other information |
Fig. 2 b is for the synoptic diagram of the application scenarios of feature matching method second embodiment of the present invention, and shown in Fig. 2 b, owing to URL is made up of host and path two parts, and the host of the URL of same stream part is identical often.Memory letter sorting unit at first extracts host name (host) 21 after receiving and needing coupling URL, adopts certain algorithm (as the hash computing) to search history feature data acquisition 22 with host as input parameter.Lookup result is adjudicated 23, if the host of current input URL exists in the history feature data acquisition, promptly the match is successful for host, if the record of the host of this URL not in the history feature data acquisition, then success of host coupling.Host is after the match is successful, memory letter sorting unit does not need the host with the URL of current input to give matching engine to carry out characteristic matching, only need the path of this URL is sent to matching engine, the while begins to mate to path in feature database position sends to matching engine 24 and carries out characteristic matching.If the not success of host coupling then can all send to matching engine with URL and carry out characteristic matching.If it is that complete URL and this URL have matched rule in feature database 26 that each in the matching engine mates data to be matched that sub-engine receives, when then the match is successful with the host of this URL and this host, host and corresponding path are begun to mate in feature database position writes in the history feature data acquisition (as the hash table) of memory letter sorting unit on the one hand, on the one hand matching result is sent to results management module 25 and handle, determine final output result; If coupling is unsuccessful, then can exports and mate not successful information.
Present embodiment compares host and the history feature data acquisition of URL, if there are history feature data in the history feature data acquisition with the host coupling, the path of this URL can be sent to matching engine mates, do not need all host to be sent to matching engine mates at every turn, therefore reduce the frequency in access characteristic storehouse, improved matching speed and efficient.
Fig. 3 a is the schematic flow sheet of feature matching method the 3rd embodiment of the present invention, and shown in Fig. 3 a, on the basis of feature matching method first embodiment of the present invention, if the input data are character string, this feature matching method comprises:
Table 2
| Character string | Matching result | Other information |
After step 302, memory letter sorting unit receive the character string of follow-up input, search the history feature data acquisition, determine that this character string is whether in the history feature data acquisition according to certain algorithm of searching.If in the history feature data acquisition, then execution in step 303 for the character string of current input; Otherwise, if the character string of current input then carries out 304 not in the history feature data acquisition.
When if there have been the history feature data of all mating in the character string of current input in the history feature data acquisition, matching engine does not need to give matching engine with this character string and mates, can with in the history feature data acquisition with this character string all the matching result of the history feature data of coupling send to the results management module, determine the output result by the results management module.
Fig. 3 b is the synoptic diagram of the application scenarios of feature matching method the 3rd embodiment of the present invention, shown in Fig. 3 b, after memory letter sorting unit receives character string, selection is searched algorithm 31 and is searched history feature data acquisition 32, lookup result is adjudicated 33, if the character string of current input exists in the history feature data acquisition, promptly the match is successful, matching result sent to results management module 35 handle; If coupling is success not, then all or part of matching engine 34 that sends to of character string is carried out characteristic matching.Each of matching engine mates sub-engine can utilize data structure in the feature database 36, character string is mated, if the match is successful, send matching result to results management module 35, and to remembering the matching result that the letter sorting unit sends this character string and character string, coupling is unsuccessful, can export and mate not successful information.
Present embodiment compares character string and the history feature data acquisition of having learnt, if there are the history feature data of all mating in the history feature data acquisition with character string, can directly obtain matching result, do not need all character string to be sent to matching engine mates at every turn, reduce the frequency in access characteristic storehouse, improved matching speed and efficient.
Fig. 4 is the structural representation of characteristic matching device first embodiment of the present invention, and as shown in Figure 4, this characteristic matching device comprises memory letter sorting unit 41 and matching engine 43; Wherein, memory letter sorting unit 41 comprises: data set matching module 411 and judging module 413.
Wherein, data set matching module 411 is used for according to the history feature data acquisition of having learnt the input data that receive being carried out characteristic matching;
Judging module 413 is used for if there are the history feature data of all mating with described input data in described history feature data acquisition, then according to obtaining matching result with the history feature data of described input Data Matching; If do not have the history feature data of all mating in the described history feature data acquisition, then all or part of matching engine 43 that is sent to of described input data carried out characteristic matching with described input data;
Wherein, feature database can be stored in the characteristic matching device, also can be stored in the chip external memory.Matching engine 43 can comprise that one or more serial or parallels mate sub-engine, after receiving the input data, each mates sub-engine access characteristic storehouse, by certain algorithm for example the hash algorithm rule that can determine to import in data and the feature database whether mate, if, then each mates sub-engine the characteristic and the matching result of coupling is sent to memory letter sorting unit 41, memory letter sorting unit 41 is saved in the history feature data acquisition this characteristic as the history feature data, for follow-up comparison.
In the follow-up processing procedure, after memory letter sorting unit 41 receives the input data, data set matching module 411 will import data with before learnt with feature database in the regular history feature data that are complementary compare.If there are the history feature data of all mating with described input data in the history feature data acquisition, then judging module 413 is according to exporting corresponding matching result with the history feature data of input Data Matching; If there are not the history feature data of all mating in the history feature data acquisition with described input data, then judging module 413 can be carried out characteristic matching by matching engine 43 according to the rule of storing in the feature database with all or part of matching engine 43 that is sent to of input data.
Present embodiment memory letter sorting unit can compare input data and the history feature data acquisition of having learnt, if there are the history feature data of all mating in the history feature data acquisition with the input data, can directly obtain matching result, do not need all will import data sends to matching engine and mates at every turn, reduce the frequency in access characteristic storehouse, improved matching speed and efficient.
Fig. 5 is the synoptic diagram of characteristic matching device first embodiment of the present invention, and as shown in Figure 5, on the basis of characteristic matching device first embodiment of the present invention, memory letter sorting unit 41 also comprises:
History feature data acquisition 415 is used to store history feature data and the corresponding matching result thereof that described matching engine sends.This history feature data acquisition can be represented with any data structure easy-to-look-up and record, include but not limited to the hash table.
Further, judging module 413 can comprise:
URL(uniform resource locator) submodule 51, being used for working as described input data is URL(uniform resource locator), when described URL(uniform resource locator) comprises host name and path, if history feature data in the described history feature data acquisition and described host name and path all do not match, then described host name and path are sent to matching engine and carry out characteristic matching; Perhaps, if there are the history feature data of all mating in the described history feature data acquisition with described host name, described path is sent to matching engine carries out characteristic matching, when perhaps being the history feature data of part coupling in existence and described path, the not part of coupling in described path is sent to matching engine and carries out characteristic matching, perhaps host name and described path are sent to matching engine and carry out characteristic matching; Perhaps,
In addition, because when carrying out characteristic matching, in the characteristic that may have many couplings for a certain input data, need for example handle each characteristic of coupling: choose priority the highest or the time is up-to-date as output data etc., therefore, this characteristic matching device can also comprise: results management module 45, be used to receive the matching result that unit 41 and matching engine 43 inputs are sorted in memory, and determine to export the result according to setting rule according to described matching result.
Particularly, matching engine 43 can comprise that one or more serial or parallels for example mate sub-engine: mate sub-engine 1, the sub-engine 2 of coupling etc., each mates sub-engine can have certain coupled relation between the sub-engine of separate or any a plurality of couplings, handles the back as a sub-engine of coupling and mates sub-engine continuation processing to another.After matching engine 43 receives the input data, each mates sub-engine access characteristic storehouse, by certain algorithm for example: the hash algorithm, whether the rule that can determine to import in data and the feature database mates, if, then on the one hand each mates sub-engine the characteristic and the matching result of coupling is sent to memory letter sorting unit 41, and memory letter sorting unit 41 these characteristics are saved in the history feature data acquisition as the history feature data, for follow-up comparison; The matching result that on the other hand each is mated sub-engine sends to results management module 45, results management module 45 is according to each definite final output result such as priority of mating the matching result of sub-engine, and the certain algorithm calculated result of matching result process of respectively mating sub-engine that perhaps will intercouple is as the output result.After memory letter sorting unit 41 receives the input data, data set matching module 411 is searched each bar history feature data of history feature data acquisition 415 storages, whether judging module 413 relatively obtains existing in the history feature data acquisition 415 and the identical history feature data of input data, if, then will import the corresponding matching result of the identical history feature data of data and send to results management module 45 with this, after results management module 45 is handled, export final result.If do not exist in the history feature data acquisition 415 and the identical history feature data of input data, then these input data can be sent to matching engine 43 and carry out characteristic matching; Perhaps also the part of not mating in these input data can be sent to matching engine 43 and carry out characteristic matching.
When the input data are URL(uniform resource locator) (URL), URL(uniform resource locator) submodule 51 can compare the history feature data whether the history feature data acquisition exists the host name (host) with URL to be complementary, if have, again the path (path) of URL is sent to matching engine and carries out characteristic matching; The host name that also can work as URL is when the history feature data whether the history feature data acquisition mates, also compare the path, when the path is also identical, directly send matching result to results management module 45, the path is not simultaneously, with whole paths or not the part path of coupling be sent to matching engine and carry out characteristic matching, matching engine 43 is informed in the data that memory letter sorting unit further will need to mate begin to mate in feature database position.
When described input data are character string, character string submodule 53 can compare the history feature data acquisition and whether have the history feature data of all mating with character string, if have, send the matching result of the history feature data that match to results management module 45; If do not have, can also more whether be present in the history feature data that character string is partly mated, if there are the history feature data of part coupling, then both character string directly all can be sent to matching engine 43 and carry out characteristic matching, also the part of character string can not mated sends to matching engine 43 and carries out characteristic matching, and matching engine 43 is informed in the data that memory letter sorting unit further will need to mate begin to mate in feature database position.
Present embodiment memory letter sorting unit can compare input data and the history feature data acquisition of having learnt, if there are the history feature data of all mating in the history feature data acquisition with the input data, can directly obtain matching result, do not need all will import data sends to matching engine and mates at every turn, reduce the frequency in access characteristic storehouse, improved matching speed and efficient.
One of ordinary skill in the art will appreciate that: all or part of step that realizes said method embodiment can be finished by the relevant hardware of programmed instruction, aforesaid program can be stored in the computer read/write memory medium, this program is carried out the step that comprises said method embodiment when carrying out; And aforesaid storage medium comprises: various media that can be program code stored such as ROM, RAM, magnetic disc or CD.
It should be noted that at last: above embodiment only in order to technical scheme of the present invention to be described, is not intended to limit; Although with reference to previous embodiment the present invention is had been described in detail, those of ordinary skill in the art is to be understood that: it still can be made amendment to the technical scheme that aforementioned each embodiment put down in writing, and perhaps part technical characterictic wherein is equal to replacement; And these modifications or replacement do not make the essence of appropriate technical solution break away from the scope of various embodiments of the present invention technical scheme.
Claims (8)
1. a feature matching method is characterized in that, comprising:
According to the history feature data acquisition of having learnt, the input data that receive are carried out characteristic matching;
If there are the history feature data of all mating in the described history feature data acquisition, then according to obtaining matching result with the history feature data of described input Data Matching with described input data;
If do not have the history feature data of all mating in the described history feature data acquisition, then all or part of matching engine that is sent to of described input data carried out characteristic matching with described input data.
2. feature matching method according to claim 1, it is characterized in that, described input data are URL(uniform resource locator), described URL(uniform resource locator) comprises host name and path, described if there are not the history feature data of all mating in the described history feature data acquisition with described input data, then all or part of matching engine that is sent to of described input data is carried out characteristic matching, comprising:
If history feature data in the described history feature data acquisition and described host name and path all do not match, then described host name and path are sent to matching engine and carry out characteristic matching;
Perhaps,
If have the history feature data of all mating in the described history feature data acquisition, described path be sent to matching engine carry out characteristic matching with described host name; Perhaps exist when be the history feature data of part coupling, the part of not mating in described path is sent to matching engine carries out characteristic matching with described path; Perhaps host name and described path are sent to matching engine and carry out characteristic matching.
3. feature matching method according to claim 1, it is characterized in that, described input data are character string, described if there are not the history feature data of all mating in the described history feature data acquisition with described input data, then all or part of matching engine that is sent to of described input data is carried out characteristic matching, comprising:
If history feature data and described character string in the described history feature data acquisition all do not match, then described character string is sent to matching engine and carries out characteristic matching; Perhaps,
If do not have the history feature data of all mating in the described history feature data acquisition, but have the history feature data of partly mating, then described character string is sent to matching engine and carries out characteristic matching with described character string with described character string; Perhaps the part that described character string is not mated is sent to matching engine and carries out characteristic matching.
4. according to the arbitrary described feature matching method of claim 1-3, it is characterized in that, also comprise:
Receive and learn history feature data and corresponding matching result thereof that described matching engine sends.
5. a characteristic matching device is characterized in that, comprising: memory letter sorting unit and matching engine; Described memory letter sorting unit comprises: data set matching module and judging module;
Described data set matching module is used for according to the history feature data acquisition of having learnt the input data that receive being carried out characteristic matching;
Described judging module is used for if there are the history feature data of all mating with described input data in described history feature data acquisition, then according to obtaining matching result with the history feature data of described input Data Matching; If do not exist in the described history feature data acquisition and the whole history feature data of coupling of described input data, then described input data all or part of is sent to described matching engine and carries out characteristic matching;
Described matching engine is used for according to feature database the input data that receive being carried out characteristic matching.
6. characteristic matching device according to claim 5 is characterized in that, described memory letter sorting unit also comprises:
The history feature data acquisition is used to store history feature data and the corresponding matching result thereof that described matching engine sends.
7. according to claim 5 or 6 described characteristic matching devices, it is characterized in that described judging module comprises:
The URL(uniform resource locator) submodule, being used for working as described input data is URL(uniform resource locator), when described URL(uniform resource locator) comprises host name and path:, then described host name and path are sent to matching engine and carry out characteristic matching if the history feature data in the described history feature data acquisition and described host name and path all do not match; Perhaps, if there are the history feature data of all mating in the described history feature data acquisition with described host name, described path is sent to matching engine carries out characteristic matching, when perhaps being the history feature data of part coupling in existence and described path, the not part of coupling in described path is sent to matching engine and carries out characteristic matching, perhaps host name and described path are sent to matching engine and carry out characteristic matching;
Perhaps,
The character string submodule is used for when described input data are character string, if history feature data and described character string in the described history feature data acquisition all do not match, then described character string is sent to matching engine and carries out characteristic matching; Perhaps if there are not the history feature data of all mating in the described history feature data acquisition with described character string, but there are the history feature data of partly mating with described character string, then described character string is sent to matching engine and carries out characteristic matching, perhaps the part that described character string is not mated is sent to matching engine and carries out characteristic matching.
8. according to claim 5 or 6 described characteristic matching devices, it is characterized in that, also comprise:
The results management module is used to receive the matching result that unit and the input of described matching engine are sorted in described memory, determines to export the result according to setting rule according to described matching result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010101271638A CN102193948A (en) | 2010-03-16 | 2010-03-16 | Feature matching method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010101271638A CN102193948A (en) | 2010-03-16 | 2010-03-16 | Feature matching method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102193948A true CN102193948A (en) | 2011-09-21 |
Family
ID=44602027
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010101271638A Pending CN102193948A (en) | 2010-03-16 | 2010-03-16 | Feature matching method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102193948A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103248609A (en) * | 2012-02-06 | 2013-08-14 | 同方股份有限公司 | System, device and method for detecting data from end to end |
| CN103685320A (en) * | 2013-12-31 | 2014-03-26 | 北京网康科技有限公司 | Feature matching method and device of network data package |
| CN104121916A (en) * | 2013-04-27 | 2014-10-29 | 国际商业机器公司 | Method and system for map matching |
| CN104239565A (en) * | 2014-09-28 | 2014-12-24 | 陆嘉恒 | Name automatic prompting method based on academic research |
| CN105897811A (en) * | 2015-01-26 | 2016-08-24 | 中国移动通信集团公司 | data synchronization method and device |
| CN106484774A (en) * | 2016-09-12 | 2017-03-08 | 北京歌华有线电视网络股份有限公司 | A kind of correlating method of multisource video metadata and system |
| CN106503118A (en) * | 2016-10-18 | 2017-03-15 | 国云科技股份有限公司 | A kind of data acquisition system and its implementation based on HC TABLE |
| CN107404486A (en) * | 2017-08-04 | 2017-11-28 | 厦门市美亚柏科信息股份有限公司 | Parse method, apparatus, terminal device and the storage medium of Http data |
| CN108388676A (en) * | 2018-03-27 | 2018-08-10 | 广东工业大学 | A kind of mold data matching process, apparatus and system based on simulated annealing |
| CN110188156A (en) * | 2019-06-04 | 2019-08-30 | 国家电网有限公司 | A kind of work transmission line three dimensional design achievement key message extracting method and system |
| CN113722621A (en) * | 2021-08-30 | 2021-11-30 | 康键信息技术（深圳）有限公司 | Service processing method, device and storage medium based on URL |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002024117A (en) * | 2000-05-08 | 2002-01-25 | Nternet Number Corp | Method and system for accessing information on network |
| CN101567006A (en) * | 2009-05-25 | 2009-10-28 | 中兴通讯股份有限公司 | Database system and distributed SQL statement execution plan reuse method |
| CN101605129A (en) * | 2009-06-23 | 2009-12-16 | 北京理工大学 | A kind of URL lookup method that is used for the url filtering system |
-
2010
- 2010-03-16 CN CN2010101271638A patent/CN102193948A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002024117A (en) * | 2000-05-08 | 2002-01-25 | Nternet Number Corp | Method and system for accessing information on network |
| CN101567006A (en) * | 2009-05-25 | 2009-10-28 | 中兴通讯股份有限公司 | Database system and distributed SQL statement execution plan reuse method |
| CN101605129A (en) * | 2009-06-23 | 2009-12-16 | 北京理工大学 | A kind of URL lookup method that is used for the url filtering system |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103248609A (en) * | 2012-02-06 | 2013-08-14 | 同方股份有限公司 | System, device and method for detecting data from end to end |
| CN104121916A (en) * | 2013-04-27 | 2014-10-29 | 国际商业机器公司 | Method and system for map matching |
| CN103685320A (en) * | 2013-12-31 | 2014-03-26 | 北京网康科技有限公司 | Feature matching method and device of network data package |
| CN104239565A (en) * | 2014-09-28 | 2014-12-24 | 陆嘉恒 | Name automatic prompting method based on academic research |
| CN105897811A (en) * | 2015-01-26 | 2016-08-24 | 中国移动通信集团公司 | data synchronization method and device |
| CN105897811B (en) * | 2015-01-26 | 2019-04-23 | 中国移动通信集团公司 | A kind of method of data synchronization and device |
| CN106484774A (en) * | 2016-09-12 | 2017-03-08 | 北京歌华有线电视网络股份有限公司 | A kind of correlating method of multisource video metadata and system |
| CN106484774B (en) * | 2016-09-12 | 2020-10-20 | 北京歌华有线电视网络股份有限公司 | Correlation method and system for multi-source video metadata |
| CN106503118B (en) * | 2016-10-18 | 2019-06-21 | 国云科技股份有限公司 | A kind of data acquisition system and its implementation based on HC-TABLE |
| CN106503118A (en) * | 2016-10-18 | 2017-03-15 | 国云科技股份有限公司 | A kind of data acquisition system and its implementation based on HC TABLE |
| CN107404486B (en) * | 2017-08-04 | 2020-05-22 | 厦门市美亚柏科信息股份有限公司 | Method, device, terminal equipment and storage medium for analyzing Http data |
| CN107404486A (en) * | 2017-08-04 | 2017-11-28 | 厦门市美亚柏科信息股份有限公司 | Parse method, apparatus, terminal device and the storage medium of Http data |
| CN108388676A (en) * | 2018-03-27 | 2018-08-10 | 广东工业大学 | A kind of mold data matching process, apparatus and system based on simulated annealing |
| CN110188156A (en) * | 2019-06-04 | 2019-08-30 | 国家电网有限公司 | A kind of work transmission line three dimensional design achievement key message extracting method and system |
| CN113722621A (en) * | 2021-08-30 | 2021-11-30 | 康键信息技术（深圳）有限公司 | Service processing method, device and storage medium based on URL |
| CN113722621B (en) * | 2021-08-30 | 2023-11-14 | 康键信息技术（深圳）有限公司 | URL-based service processing method, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102193948A (en) | Feature matching method and device | |
| CN101782919B (en) | Web form data output method, device and form processing system | |
| CN102857493B (en) | Content filtering method and device | |
| CN102148805B (en) | Feature matching method and device | |
| CN103136228A (en) | Image search method and image search device | |
| WO2008043645B1 (en) | Establishing document relevance by semantic network density | |
| CN102082762A (en) | Protocol identification method and device and system for same | |
| CN108182523A (en) | The treating method and apparatus of fault data, computer readable storage medium | |
| CN104021161A (en) | Cluster storage method and device | |
| US8930389B2 (en) | Mutual search and alert between structured and unstructured data stores | |
| CN102870116B (en) | Method and apparatus for content matching | |
| CN105677683A (en) | Batch data query method and device | |
| CN102754394A (en) | Method for hash table storage, method for hash table lookup, and devices thereof | |
| CN102780681A (en) | URL (Uniform Resource Locator) filtering system and URL filtering method | |
| CN104298736A (en) | Method and device for aggregating and connecting data as well as database system | |
| CN105335402A (en) | Search method, index data generation method and device on the basis of static Cache | |
| CN102437937A (en) | Deep packet inspection method | |
| CN111666468A (en) | Method for searching personalized influence community in social network based on cluster attributes | |
| CN110263021B (en) | Theme library generation method based on personalized label system | |
| US8756093B2 (en) | Method of monitoring a combined workflow with rejection determination function, device and recording medium therefor | |
| CN106933919A (en) | The connection method of tables of data and device | |
| CN114598597A (en) | Multi-source log analysis method and device, computer equipment and medium | |
| CN104424316A (en) | Data storage method, data searching method, related device and system | |
| CN113641742A (en) | Data extraction method, device, equipment and storage medium | |
| US20170308574A1 (en) | Method and apparatus for reducing query processing time by dynamically changing algorithms and computer readable medium therefor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20110921 |