CN105389722A - Malicious order identification method and device - Google Patents

Abstract

The invention relates to a malicious order identification method and device and belongs to the electronic technology application field. The method comprises the steps of: according a preset first word segmentation algorithm, carrying out word segmentation on an address to be identified, and obtaining words to be identified in the address to be identified; counting the number of words belonging to a malicious word database in the words to be identified, wherein the malicious word database is established in advance, and at least one malicious word used for identifying malicious addresses is recorded in the malicious word database; according to the number of words belonging to the malicious word database, judging whether the address to be identified is a malicious address; and when the address to be identified is the malicious address, determining that an order corresponding to the address to be identified is a malicious order. By adopting the malicious order identification method and device, the accuracy of malicious order identification is improved, and the problem in the prior art that the accuracy is relatively low when the malicious order is identified by means of similarity. The malicious order identification method and device are used for identifying malicious orders.

Description

Malice order recognition methods and device

Technical field

The disclosure relates to application of electronic technology field, particularly one malice order recognition methods and device.

Background technology

Along with the fast development of e-commerce technology, various marketing methods emerge in an endless stream, the marketing methods of popular a kind of panic buying instantly, large short pattern, such as: be lower price by merchandise valuation, and buy time point of specifying is open.In this case, some malicious users may be there are, adopt the mode running counter to active rule, preempting resources in enormous quantities, then sell with high price.The behavior of these malicious users has had a strong impact on the interests that other have the user of true buying intention.

In correlation technique, when malicious user carries out large batch of buying behavior in electric business's platform, in the sequence information of this malicious user stored in electricity business platform database, a large amount of duplicate messages may be there is: such as address information, telephone number, results people name or place an order time the Internet protocol of terminal that uses (be called for short: IP) address etc.Therefore, in correlation technique, generally identify malicious user by carrying out similarity evaluation to sequence information.Such as can calculate the similarity of the address information in each order, and order address information similarity being exceeded certain threshold value is defined as alternative order, if the quantity of this alternative order also exceedes certain threshold value, then this alternative order can be defined as malice order.

Summary of the invention

In order to solve the problem in correlation technique, present disclose provides the order recognition methods of a kind of malice and device.Described technical scheme is as follows:

According to the first aspect of disclosure embodiment, provide the recognition methods of a kind of malice order, described method comprises:

According to the first participle algorithm preset, treat identification address and carry out participle, obtain the word to be identified in described address to be identified;

Add up in described word to be identified, belong to the number of the word in malice word storehouse, described malice word storehouse is what set up in advance, and described malice word storehouse records at least one for identifying the malice word of malice address;

According to the number of the described word belonged in described malice word storehouse, judge whether described address to be identified is malice address;

When described address to be identified is malice address, determine that order corresponding to described address to be identified is for malice order.

Optionally, described in described basis, belong to the number of the word in described malice word storehouse, judge whether described address to be identified is malice address, comprising:

Add up the number n of the word to be identified in described address to be identified;

According to the number n of the word to be identified in the number m of the described word belonged in described malice word storehouse and described address to be identified, calculated the malice degree scoring S1 of described address to be identified by address malice degree evaluate formula, described address malice degree evaluate formula is: S1=m/n;

Judge whether the malice degree scoring S1 of described address to be identified is greater than predetermined threshold value t;

When described malice degree scoring S1 is greater than predetermined threshold value t, determine that described address to be identified is for malice address;

When described malice degree scoring S1 is not more than predetermined threshold value t, determine that described address to be identified is not malice address.

Optionally, described method also comprises: obtain set of words to be identified;

For the arbitrary word in described set of words to be identified, calculate the malice degree scoring S2 of described arbitrary word according to word malice degree evaluate formula, described word malice degree evaluate formula is:

Wherein, abs () expression takes absolute value to content in bracket, k1 is default general character constant, k2 is default length characteristic value, tf is described arbitrary word word frequency in a database, df is described arbitrary word document frequency in a database, and L is containing the number comprising the different address of number of characters in the address of described arbitrary word in described database;

Judge whether the malice degree scoring S2 of described arbitrary word is less than default Evaluation threshold k3;

When the malice degree scoring S2 of described arbitrary word is less than default Evaluation threshold k3, determine that described arbitrary word is for malice word;

According to the malice word in described set of words to be identified, set up described malice word storehouse.

Optionally, described acquisition set of words to be identified, comprising:

According to the second segmentation methods preset, participle is carried out to each address in database;

By the described set of words to be identified of word composition after participle.

Optionally, described second segmentation methods comprises described first participle algorithm, and described second segmentation methods comprises at least one in two character segmentation algorithms and three character segmentation algorithms.

According to the second aspect of disclosure embodiment, provide a kind of malice order recognition device, described device comprises:

Word-dividing mode, is configured to, according to the first participle algorithm preset, treat identification address and carry out participle, obtain the word to be identified in described address to be identified;

Statistical module, be configured in the described word to be identified of statistics, belong to the number of the word in malice word storehouse, described malice word storehouse is what set up in advance, and described malice word storehouse records at least one for identifying the malice word of malice address;

First judge module, is configured to the number belonging to the word in described malice word storehouse described in basis, judges whether described address to be identified is malice address;

First determination module, is configured to when described address to be identified is for malice address, determines that order corresponding to described address to be identified is for malice order.

Optionally, described first judge module, is configured to:

Add up the number n of the word to be identified in described address to be identified;

According to the number n of the word to be identified in the number m of the described word belonged in described malice word storehouse and described address to be identified, calculated the malice degree scoring S1 of described address to be identified by address malice degree evaluate formula, described address malice degree evaluate formula is: S1=m/n;

Judge whether the malice degree scoring S1 of described address to be identified is greater than predetermined threshold value t;

When described malice degree scoring S1 is greater than predetermined threshold value t, determine that described address to be identified is for malice address;

When described malice degree scoring S1 is not more than predetermined threshold value t, determine that described address to be identified is not malice address.

Optionally, described device also comprises:

Acquisition module, is configured to obtain set of words to be identified;

Computing module, is configured to for the arbitrary word in described set of words to be identified, and calculate the malice degree scoring S2 of described arbitrary word according to word malice degree evaluate formula, described word malice degree evaluate formula is: $S2=\frac{abs(tf-df)}{k1}+\frac{L}{k2};$

Wherein, abs () expression takes absolute value to content in bracket, k1 is default general character constant, k2 is default length characteristic value, tf is described arbitrary word word frequency in a database, df is described arbitrary word document frequency in a database, and L is containing the number comprising the different address of number of characters in the address of described arbitrary word in described database;

Second judge module, is configured to judge whether the malice degree scoring S2 of described arbitrary word is less than default Evaluation threshold k3;

Second determination module, is configured to when the malice degree scoring S2 of described arbitrary word is less than default Evaluation threshold k3, determines that described arbitrary word is for malice word;

Set up module, be configured to according to the malice word in described set of words to be identified, set up described malice word storehouse.

Optionally, described acquisition module, is configured to:

According to the second segmentation methods preset, participle is carried out to each address in database;

By the described set of words to be identified of word composition after participle.

Optionally, described second segmentation methods comprises described first participle algorithm, and described second segmentation methods comprises at least one in two character segmentation algorithms and three character segmentation algorithms.

According to the third aspect of disclosure embodiment, provide a kind of malice order recognition device, described device comprises:

Processor;

For storing the storer of the executable instruction of described processor;

Wherein, described processor is configured to:

According to the first participle algorithm preset, treat identification address and carry out participle, obtain the word to be identified in described address to be identified;

Add up in described word to be identified, belong to the number of the word in malice word storehouse, described malice word storehouse is what set up in advance, and described malice word storehouse records at least one for identifying the malice word of malice address;

According to the number of the described word belonged in described malice word storehouse, judge whether described address to be identified is malice address;

When described address to be identified is malice address, determine that order corresponding to described address to be identified is for malice order.

The technical scheme that disclosure embodiment provides can comprise following beneficial effect:

A kind of malice order recognition methods that disclosure embodiment provides and device, according to the first participle algorithm preset, can treat identification address and carry out participle, obtain the word to be identified in described address to be identified; And add up in described word to be identified, belong to the number of the word in malice word storehouse, described malice word storehouse is what set up in advance, and described malice word storehouse records at least one for identifying the malice word of malice address; According to the number of the described word belonged in described malice word storehouse, judge whether described address to be identified is malice address; When described address to be identified is malice address, determine that order corresponding to described address to be identified is for malice order.The malice order recognition methods that disclosure embodiment provides, can be identified the address in order by the malice word storehouse set up in advance, and then determines whether this order is malice order, improves accuracy when identifying malice order.

Should be understood that, it is only exemplary that above general description and details hereinafter describe, and can not limit the disclosure.

Accompanying drawing explanation

In order to be illustrated more clearly in embodiment of the present disclosure, below the accompanying drawing used required in describing embodiment is briefly described, apparently, accompanying drawing in the following describes is only embodiments more of the present disclosure, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.

Fig. 1-1 is the schematic diagram of the implementation environment involved by the recognition methods of malice order that disclosure section Example provides;

Fig. 1-2 is the process flow diagram of a kind of malice order recognition methods according to an exemplary embodiment;

Fig. 2-1 is the process flow diagram of the another kind of malice order recognition methods according to an exemplary embodiment;

Fig. 2-2 is a kind of method flow diagrams setting up malice word storehouse according to an exemplary embodiment;

Fig. 3-1 is the block diagram of a kind of malice order recognition device according to an exemplary embodiment;

Fig. 3-2 is block diagrams of the another kind of malice order recognition device according to an exemplary embodiment;

Fig. 4 is the block diagram of another the malice order recognition device according to an exemplary embodiment.

Accompanying drawing to be herein merged in instructions and to form the part of this instructions, shows and meets embodiment of the present disclosure, and is used from instructions one and explains principle of the present disclosure.

Embodiment

In order to make object of the present disclosure, technical scheme and advantage clearly, be described in further detail the disclosure below in conjunction with accompanying drawing, obviously, described embodiment is only a part of embodiment of the disclosure, instead of whole embodiments.Based on the embodiment in the disclosure, those of ordinary skill in the art are not making other embodiments all obtained under creative work prerequisite, all belong to the scope of disclosure protection.

Fig. 1-1 is the schematic diagram of the implementation environment involved by the recognition methods of malice order that disclosure section Example provides.This implementation environment can comprise: server 110 and at least one terminal 120.Server 110 can be a station server, or the server cluster be made up of some station servers, or a cloud computing service center.Terminal 120 can be smart mobile phone, computer, multimedia player, electronic reader, Wearable device etc.Can be set up by cable network or wireless network between server 110 and terminal 120 and connect.

Wherein, user can fill in purchase order by terminal 120 in shopping platform, can store the purchase order that user fills in, and analyze this order in the server 110 corresponding to this shopping platform, judges whether this order is malice order.

Fig. 1-2 is the process flow diagram of a kind of malice order recognition methods according to an exemplary embodiment, and the method can be applied in the server 110 shown in Fig. 1-1, and as shown in Figure 1-2, the method comprises:

In a step 101, according to the first participle algorithm preset, treat identification address and carry out participle, obtain the word to be identified in this address to be identified.

In a step 102, add up in this word to be identified, belong to the number of the word in malice word storehouse, this malice word storehouse is what set up in advance, and this malice word storehouse records at least one for identifying the malice word of malice address.

In step 103, according to the number that this belongs to the word in this malice word storehouse, judge whether this address to be identified is malice address.

At step 104, when this address to be identified is malice address, determine that order corresponding to this address to be identified is for malice order.

In sum, a kind of malice order recognition methods that embodiment of the present disclosure provides, according to the first participle algorithm preset, can treat identification address and carries out participle, obtain the word to be identified in described address to be identified; And add up in described word to be identified, belong to the number of the word in malice word storehouse; According to the number of the described word belonged in described malice word storehouse, judge whether described address to be identified is malice address; When described address to be identified is malice address, determine that order corresponding to described address to be identified is for malice order.The malice order recognition methods that disclosure embodiment provides, can be identified the address in order by the malice word storehouse set up in advance, and then determines whether this order is malice order, and the accuracy of this malice order recognition methods is higher.

Optionally, this belongs to the number of the word in this malice word storehouse according to this, judges whether this address to be identified is malice address, comprising:

Add up the number n of the word to be identified in this address to be identified;

Belong to the number n of the word to be identified in the number m of the word in this malice word storehouse and this address to be identified according to this, calculated the malice degree scoring S1 of this address to be identified by address malice degree evaluate formula, this address malice degree evaluate formula is: S1=m/n;

Judge whether the malice degree scoring S1 of this address to be identified is greater than predetermined threshold value t;

When this malice degree scoring S1 is greater than predetermined threshold value t, determine that this address to be identified is for malice address;

When this malice degree scoring S1 is not more than predetermined threshold value t, determine that this address to be identified is not malice address.

Optionally, the method also comprises: obtain set of words to be identified;

For the arbitrary word in this set of words to be identified, calculate the malice degree scoring S2 of this arbitrary word according to word malice degree evaluate formula, this word malice degree evaluate formula is:

Wherein, abs () expression takes absolute value to content in bracket, k1 is default general character constant, k2 is default length characteristic value, tf is this arbitrary word word frequency in a database, df is this arbitrary word document frequency in a database, and L is containing the number comprising the different address of number of characters in the address of this arbitrary word in this database;

Judge whether the malice degree scoring S2 of described arbitrary word is less than default Evaluation threshold k3;

When the malice degree scoring S2 of this arbitrary word is less than default Evaluation threshold k3, determine that this arbitrary word is for malice word;

According to the malice word in this set of words to be identified, set up this malice word storehouse.

Optionally, this acquisition set of words to be identified, comprising:

According to the second segmentation methods preset, participle is carried out to each address in database;

Word after participle is formed this set of words to be identified.

Optionally, this second segmentation methods comprises this first participle algorithm, and this second segmentation methods comprises at least one in two character segmentation algorithms and three character segmentation algorithms.

In sum, a kind of malice order recognition methods that embodiment of the present disclosure provides, according to the first participle algorithm preset, can treat identification address and carries out participle, obtain the word to be identified in described address to be identified; And add up in described word to be identified, belong to the number of the word in malice word storehouse; According to the number of the described word belonged in described malice word storehouse, judge whether described address to be identified is malice address; When described address to be identified is malice address, determine that order corresponding to described address to be identified is for malice order.The malice order recognition methods that disclosure embodiment provides, can be identified the address in order by the malice word storehouse set up in advance, and then determines whether order is malice order, and the accuracy of this malice order recognition methods is higher.

Fig. 2-1 is the process flow diagram of the another kind of malice order recognition methods according to an exemplary embodiment, and the method can be applied in the server 110 shown in Fig. 1-1, and as shown in Fig. 2-1, the method comprises:

In step 201, according to the first participle algorithm preset, treat identification address and carry out participle, obtain the word to be identified in this address to be identified.Perform step 202.

In the disclosed embodiments, when server needs to judge whether a certain order is malice order, the address that this order can be comprised is as address to be identified, and according to the first participle algorithm preset, carry out participle to this address to be identified, wherein this first participle algorithm preset can be two character segmentation algorithms or three character segmentation algorithms.Two character segmentation algorithms and three character segmentation algorithms refer to that treating identification address carries out order cutting according to every two or every three individual characters, and then obtain word to be identified.When carrying out two words or three character segmentations, continuous print numeral or letter can be defined as an individual character, the word such as, obtained after carrying out three character segmentations to " scientific and technological road 20A seat " can be: scientific and technological road, skill road 20, road 20A, 20A seat.Wherein, using continuous number " 20 " and letter " A " as an individual character.

Further, in order to improve the precision to Address Recognition, treating before identification address carries out participle, first can also remove the stop words in address to be identified, this stop words can treat some less characters of identification address impact, such as punctuation mark, space and other special characters etc. for what pre-set.

Example, suppose that address to be identified in a certain order is for " the former new micro-purport in Jie Duran township out-of-the-way several days No. 2 ", then after carrying out participle according to two character segmentation algorithms to this address to be identified, the word to be identified obtained can be: street tripe, and tripe is right, right township, township is former, newly former, newly micro-, micro-purport, purport is out-of-the-way, out-of-the-way several, several days, they 2, No. 2.

In step 202., add up in this word to be identified, belong to the number m of the word in malice word storehouse.Perform step 203.

In the disclosed embodiments, can set up a malice word storehouse in server in advance, this malice word storehouse records at least one for identifying the malice word of malice address.Whether after server determines word to be identified, can detect this word to be identified is the malice word stored in this malice word storehouse, and adds up in word to be identified the number m of the word belonged in malice word storehouse.

Fig. 2-2 is a kind of process flow diagrams setting up the method in malice word storehouse according to an exemplary embodiment, and as shown in Fig. 2-2, the method comprises:

In step 2021, according to the second segmentation methods preset, participle is carried out to each address in database.Perform step 2022.

In the disclosed embodiments, when the quantity on order stored in server database reaches certain threshold value, according to the second segmentation methods preset, participle can be carried out to the address that each order in this database comprises, to set up malice word storehouse.For the ease of the word stored in word to be identified and malice word storehouse is contrasted, this second segmentation methods preset needs to comprise this first participle algorithm, this second segmentation methods can be at least one in two character segmentation algorithms and three character segmentation algorithms, namely this second segmentation methods can be two character segmentation algorithms, also can be three character segmentation algorithms, or can also be the combination algorithm of two character segmentation algorithms and three character segmentation algorithms.If only include a kind of segmentation algorithm in this second segmentation methods, then this first participle algorithm is identical with this second segmentation methods, if this second segmentation methods comprises at least two kinds of segmentation algorithm, then this first participle algorithm can be identical with this second segmentation methods, the arbitrary segmentation algorithm at least two kinds of segmentation algorithm that also can comprise for this second segmentation methods.

Example, the address in the order stored in tentation data storehouse comprises: " the former new micro-purport in Jie Duran township out-of-the-way several days No. 2 " and " the crafty round Piao Tian of nine new sciences and technologies No. 2 A ", then carry out the result after participle according to two character segmentation algorithms to address " the former new micro-purport in Jie Duran township out-of-the-way several days No. 2 " to be: { street tripe, tripe is right, right township, and township is former, newly former, newly micro-, micro-purport, purport is out-of-the-way, out-of-the-way several, several days, sky 2, No. 2 }, can be namely 12 words by this address cutting by two character segmentation algorithms; Carrying out the result after participle to address " the crafty round Piao Tian of nine new sciences and technologies No. 2 A " is: { nine, nine is new, and newly, science and technology, skill is crafty, crafty circle, circle Piao, Piao Tian, sky 2, No. 2, number A}, can be namely 11 words by this address cutting by two character segmentation algorithms.

In step 2022, the word after participle is formed set of words to be identified.Perform step 2023.

After participle is carried out to each address in database, the word after all addresses participle in this database can be formed set of words to be identified.Example, after participle is carried out to address " the former new micro-purport in Jie Duran township out-of-the-way several days No. 2 " and " the nine crafty Yuan PiaotianAZuo roads of new science and technology ", according to the set of words to be identified that word segmentation result forms can be: street tripe, tripe is right, right township, township is former, newly former, newly micro-, micro-purport, purport is out-of-the-way, out-of-the-way several, several days, they 2, No. 2, nine, nine is new, newly, science and technology, skill is crafty, crafty circle, circle Piao, Piao Tian, number A}.

In step 2023, for the arbitrary word in this set of words to be identified, calculate the malice degree scoring S2 of this arbitrary word according to word malice degree evaluate formula.Perform step 2024.

This word malice degree evaluate formula is:

Wherein, abs () represents and takes absolute value to content in bracket, and k1 is default general character constant, and k2 is default length characteristic value, and tf is this word word frequency in a database, i.e. the number of times that occurs in all addresses of database purchase of this word; Df is this word document frequency in a database, the number of the address namely containing this word in database; L is containing the number comprising the different address of number of characters in the address of this word in database.

In the disclosed embodiments, the general character constant k1 preset and the length characteristic value k2 preset can be arranged according to the actual conditions of the address stored in database, the corpus that also can mark with reference to some, in conjunction with above-mentioned word malice degree evaluate formula, parameter training is carried out by the mode of machine learning, and then the length characteristic value k2 determining this general character constant k1 preset and preset.

Example, suppose that the general character constant k1 that this is preset is 10, the length characteristic value k2 preset is 2, for the word " street tripe " in set of words to be identified, occur 1 time altogether in all addresses that this word " street tripe " stores in a database, therefore can determine that the word frequency of this word " street tripe " is: tf=1; Owing to comprising two addresses altogether in database, the number of the address containing this word " street tripe " is 1, therefore can determine that the document frequency of this word " street tripe " is: df=1, in database containing the number comprising the different address of number of characters in the address of this word " street tripe " be: L=1, therefore, can according to above-mentioned word malice degree evaluate formula calculate this word " street tripe " malice degree scoring be:

$S2=\frac{abs(1-1)}{10}+\frac{1}{2}=0.5$

Further, for the word " No. 2 " in set of words to be identified, can determine that this word word frequency is in a database: tf=2, document frequency is: df=2, the number of the address containing this word " No. 2 " in database is 2, and the number of characters comprised in these 2 addresses is different, therefore can determine L=2, according to above-mentioned word malice degree evaluate formula, the malice degree scoring that can calculate this word " No. 2 " is:

$S2=\frac{abs(2-2)}{10}+\frac{2}{2}=1$

It should be noted that, if adopt three character segmentation algorithms to determine set of words to be identified, the individual character number then comprised due to word to be identified is more, also larger on the impact of address style, therefore determining that this comprises the word to be identified of three individual characters, can also using the length of this word itself as a parameter in word malice degree evaluate formula, to improve the accuracy of this word malice degree evaluate formula.

In step 2024, judge whether the malice degree scoring S2 of this arbitrary word is less than default Evaluation threshold k3.

When the malice degree scoring S2 of this arbitrary word is less than default Evaluation threshold k3, perform step 2025; When the malice degree scoring S2 of this arbitrary word is not less than default Evaluation threshold k3, perform step 2026.In the disclosed embodiments, this Evaluation threshold k3 preset can be arranged according to the actual conditions of the address stored in database, the corpus that also can mark with reference to some, in conjunction with above-mentioned word malice degree evaluate formula, carry out parameter training by the mode of machine learning, and then determine this Evaluation threshold k3 preset.

Example, suppose that the Evaluation threshold k3 that this is preset is 1, then server can judge whether the malice degree scoring S2 of each word in this set of words to be identified is greater than 1, such as, because the malice degree scoring of word " street tripe " is 0.5, be less than the Evaluation threshold 1 that this is preset, then server can perform step 2025; Because the malice degree scoring of word " No. 2 " is 1, be not less than the Evaluation threshold 1 that this is preset, then server can perform step 2026

In step 2025, determine that this arbitrary word is for malice word.Perform step 2027.

When the malice degree scoring S2 of this arbitrary word is less than default Evaluation threshold k3, determine that this arbitrary word is for malice word.Example, word " street tripe " can be defined as malice word by server.

In step 2026, determine that this arbitrary word is not malice word.

When the malice degree scoring S2 of this arbitrary word is not less than default Evaluation threshold k3, determine that this arbitrary word is not malice word.Example, server can determine that word " No. 2 " is not malice word.

In step 2027, according to the malice word in this set of words to be identified, set up this malice word storehouse.

After server completes the malice degree evaluation of all words in set of words to be identified, according to the word being confirmed as malice word in this set of words to be identified, malice word storehouse can be set up.Example, suppose the method according to above-mentioned steps 2023 to step 2026, for set of words to be identified: { street tripe, tripe is right, right township, township is former, newly former, newly micro-, micro-purport, purport is out-of-the-way, out-of-the-way several, several days, they are 2 years old, No. 2, nine, nine is new, newly, science and technology, skill is crafty, crafty circle, circle Piao, Piao Tian, number A}, the malice word that server is determined is: street tripe, tripe is right, newly former, newly micro-, micro-purport, purport is out-of-the-way, out-of-the-way several, skill is crafty, crafty circle, circle Piao, Piao Tian, then server is according to above-mentioned malice word, the malice word storehouse set up can be: { street tripe, tripe is right, newly former, newly micro-, micro-purport, purport is out-of-the-way, out-of-the-way several, skill is crafty, crafty circle, circle Piao, Piao Tian }.

For above-mentioned malice word storehouse, suppose that word to be identified is: street tripe, tripe is right, right township, township is former, newly former, newly micro-, micro-purport, purport is out-of-the-way, out-of-the-way several, several days, sky 2, No. 2, then server according to malice word storehouse, can determine that the word belonged in this word to be identified in this malice word storehouse is: street tripe, tripe is right, newly former, newly micro-, micro-purport, purport is out-of-the-way and out-of-the-way several, and the number m that statistics obtains the word belonging to this malice word storehouse in this word to be identified is 7.

In step 203, the number n of the word to be identified in this address to be identified is added up.Perform step 204.

Example, the word to be identified in address to be identified " the former new micro-purport in Jie Duran township out-of-the-way several days No. 2 ": street tripe, tripe is right, right township, township is former, newly former, newly micro-, micro-purport, purport is out-of-the-way, out-of-the-way several, several days, they 2, No. 2, the number of the word to be identified that server can be determined in this address to be identified is: n=12.

In step 204, belong to the number n of the word to be identified in the number m of the word in this malice word storehouse and this address to be identified according to this, calculated the malice degree scoring S1 of this address to be identified by address malice degree evaluate formula.Perform step 205.

This address malice degree evaluate formula is: S1=m/n, and namely server can determine the malice degree scoring of this address according to the ratio in this address to be identified shared by malice word.Example, for address to be identified " the former new micro-purport in Jie Duran township out-of-the-way several days No. 2 ", because the number m belonging to the word in malice word storehouse in this address is 7, the number n of word to be identified is 12, then can determine that the malice degree scoring S1 of this address to be identified is: S1=7/12=0.583.

In step 205, judge whether the malice degree scoring S1 of this address to be identified is greater than predetermined threshold value t.

When this malice degree scoring S1 is greater than predetermined threshold value t, perform step 206; When this malice degree scoring S1 is not more than predetermined threshold value t, perform step 207.This predetermined threshold value t can set according to the actual conditions of the address stored in database, and generally, this predetermined threshold value t can be set to 0.5.Example, because the malice degree scoring S1 of address to be identified " the former new micro-purport in Jie Duran township out-of-the-way several days No. 2 " is 0.583, be greater than this predetermined threshold value 0.5, therefore server can perform step 206.

In step 206, determine that this address to be identified is for malice address.Perform step 208.

When this malice degree scoring S1 is greater than predetermined threshold value t, determine that this address to be identified is for malice address.Example, address to be identified " the former new micro-purport in Jie Duran township out-of-the-way several days No. 2 " can be defined as malice address by server.

In step 207, determine that this address to be identified is not malice address.

When this malice degree scoring S1 is not more than predetermined threshold value t, determine that this address to be identified is not malice address.Namely, when the malice word proportion in address to be identified is not more than predetermined threshold value t, server can determine that this address to be identified is not malice address.

In a step 208, determine that order corresponding to this address to be identified is for malice order.

Server determines that address to be identified is for behind malice address, can be defined as malice order, and further user corresponding for this malice order is defined as malicious user by the order corresponding to this address to be identified.Afterwards, server can perform predetermined registration operation to malicious user, such as, nullify the account of this malicious user or cancel the order etc. of this malicious user, and then ensureing the interests of other normal users.

In sum, a kind of malice order recognition methods that embodiment of the present disclosure provides, according to the first participle algorithm preset, can treat identification address and carries out participle, obtain the word to be identified in described address to be identified; And add up in described word to be identified, belong to the number of the word in malice word storehouse; According to the number of the described word belonged in described malice word storehouse, judge whether described address to be identified is malice address; When described address to be identified is malice address, determine that order corresponding to described address to be identified is for malice order.The malice order recognition methods that disclosure embodiment provides, by the malice word storehouse set up in advance, the address in order can be identified, and then determine whether this order is malice order, the accuracy of this malice order recognition methods is higher, and and malice order that similarity lower random for some, also effectively can be identified by the method for disclosure embodiment, greatly improve the accuracy rate of malice order identification.

It should be noted that, the sequencing of the step of the malice order recognition methods that disclosure embodiment provides can suitably adjust, and step also according to circumstances can carry out corresponding increase and decrease.Anyly be familiar with those skilled in the art in the technical scope that the disclosure discloses, the method changed can be expected easily, all should be encompassed within protection domain of the present disclosure, therefore repeat no more.

Fig. 3-1 is the block diagram of a kind of malice order recognition device according to an exemplary embodiment, and as shown in figure 3-1, this device comprises:

Word-dividing mode 301, is configured to, according to the first participle algorithm preset, treat identification address and carry out participle, obtain the word to be identified in this address to be identified.

Statistical module 302, is configured to add up in this word to be identified, and belong to the number of the word in malice word storehouse, this malice word storehouse is what set up in advance, and this malice word storehouse records at least one for identifying the malice word of malice address.

First judge module 303, is configured to the number belonging to the word in this malice word storehouse according to this, judges whether this address to be identified is malice address.

First determination module 304, is configured to when this address to be identified is for malice address, determines that order corresponding to this address to be identified is for malice order.

In sum, a kind of malice order recognition device that embodiment of the present disclosure provides, according to the first participle algorithm preset, can treat identification address and carries out participle, obtain the word to be identified in described address to be identified; And add up in described word to be identified, belong to the number of the word in malice word storehouse; According to the number of the described word belonged in described malice word storehouse, judge whether described address to be identified is malice address; When described address to be identified is malice address, determine that order corresponding to described address to be identified is for malice order.The malice order recognition methods that disclosure embodiment provides, can be identified the address in order by the malice word storehouse set up in advance, and then determines whether order is malice order, and the accuracy of this malice order recognition methods is higher.

Fig. 3-2 is block diagrams of the another kind of malice order recognition device according to an exemplary embodiment, and as shown in figure 3-2, this device comprises:

Word-dividing mode 301, is configured to, according to the first participle algorithm preset, treat identification address and carry out participle, obtain the word to be identified in this address to be identified.

Statistical module 302, is configured to add up in this word to be identified, and belong to the number of the word in malice word storehouse, this malice word storehouse is what set up in advance, and this malice word storehouse records at least one for identifying the malice word of malice address.

First judge module 303, is configured to the number belonging to the word in this malice word storehouse according to this, judges whether this address to be identified is malice address.

First determination module 304, is configured to when this address to be identified is for malice address, determines that order corresponding to this address to be identified is for malice order.

Acquisition module 305, is configured to obtain set of words to be identified.

Computing module 306, is configured to for the arbitrary word in this set of words to be identified, and calculate the malice degree scoring S2 of this arbitrary word according to word malice degree evaluate formula, this word malice degree evaluate formula is: $S2=\frac{abs(tf-df)}{k1}+\frac{L}{k2};$

Wherein, abs represents and takes absolute value, k1 is default general character constant, k2 is default length characteristic value, tf is this arbitrary word word frequency in a database, df is this arbitrary word document frequency in a database, and L is containing the number comprising the different address of number of characters in the address of this arbitrary word in this database.

Second judge module 307, is configured to judge whether the malice degree scoring S2 of this arbitrary word is less than default Evaluation threshold k3.

Second determination module 308, is configured to when the malice degree scoring S2 of this arbitrary word is less than default Evaluation threshold k3, determines that this arbitrary word is for malice word.

Set up module 309, be configured to according to the malice word in this set of words to be identified, set up this malice word storehouse.

Optionally, this first judge module 303, is configured to:

Add up the number n of the word to be identified in this address to be identified;

Belong to the number n of the word to be identified in the number m of the word in this malice word storehouse and this address to be identified according to this, calculated the malice degree scoring S1 of this address to be identified by address malice degree evaluate formula, this address malice degree evaluate formula is: S1=m/n;

Judge whether the malice degree scoring S1 of this address to be identified is greater than predetermined threshold value t;

When this malice degree scoring S1 is greater than predetermined threshold value t, determine that this address to be identified is for malice address;

When this malice degree scoring S1 is not more than predetermined threshold value t, determine that this address to be identified is not malice address.

Optionally, this acquisition module 305, is configured to:

According to the second segmentation methods preset, participle is carried out to each address in database;

Word after participle is formed this set of words to be identified.

Optionally, this second segmentation methods comprises this first participle algorithm, and this second segmentation methods comprises at least one in two character segmentation algorithms and three character segmentation algorithms.

In sum, a kind of malice order recognition device that embodiment of the present disclosure provides, according to the first participle algorithm preset, can treat identification address and carries out participle, obtain the word to be identified in described address to be identified; And add up in described word to be identified, belong to the number of the word in malice word storehouse; According to the number of the described word belonged in described malice word storehouse, judge whether described address to be identified is malice address; When described address to be identified is malice address, determine that order corresponding to described address to be identified is for malice order.The malice order recognition methods that disclosure embodiment provides, can be identified the address in order by the malice word storehouse set up in advance, and then determines whether order is malice order, and the accuracy of this malice order recognition methods is higher.

About the device in above-described embodiment, wherein the concrete mode of modules executable operations has been described in detail in about the embodiment of the method, will not elaborate explanation herein.

Fig. 4 is the block diagram of another the malice order recognition device 400 according to an exemplary embodiment.Such as, device 400 may be provided in a server.With reference to Fig. 4, device 400 comprises processing components 422, and it comprises one or more processor further, and the memory resource representated by storer 432, such as, for storing the instruction that can be performed by processing element 422, application program.The application program stored in storer 432 can comprise each module corresponding to one group of instruction one or more.In addition, processing components 422 is configured to perform instruction, and to perform the recognition methods of above-mentioned malice order, described method comprises:

According to the first participle algorithm preset, treat identification address and carry out participle, obtain the word to be identified in this address to be identified;

Add up in this word to be identified, belong to the number of the word in malice word storehouse, this malice word storehouse is what set up in advance, and this malice word storehouse records at least one for identifying the malice word of malice address;

According to the number that this belongs to the word in this malice word storehouse, judge whether this address to be identified is malice address;

When this address to be identified is malice address, determine that order corresponding to this address to be identified is for malice order.

Optionally, this belongs to the number of the word in this malice word storehouse according to this, judges whether this address to be identified is malice address, comprising:

Add up the number n of the word to be identified in this address to be identified;

Belong to the number n of the word to be identified in the number m of the word in this malice word storehouse and this address to be identified according to this, calculated the malice degree scoring S1 of this address to be identified by address malice degree evaluate formula, this address malice degree evaluate formula is: S1=m/n;

Judge whether the malice degree scoring S1 of this address to be identified is greater than predetermined threshold value t;

When this malice degree scoring S1 is greater than predetermined threshold value t, determine that this address to be identified is for malice address;

When this malice degree scoring S1 is not more than predetermined threshold value t, determine that this address to be identified is not malice address.

Optionally, the method also comprises: obtain set of words to be identified;

For the arbitrary word in this set of words to be identified, calculate the malice degree scoring S2 of this arbitrary word according to word malice degree evaluate formula, this word malice degree evaluate formula is:

Wherein, abs () expression takes absolute value to content in bracket, k1 is default general character constant, k2 is default length characteristic value, tf is this arbitrary word word frequency in a database, df is this arbitrary word document frequency in a database, and L is containing the number comprising the different address of number of characters in the address of this arbitrary word in this database;

Judge whether the malice degree scoring S2 of this arbitrary word is less than default Evaluation threshold k3;

When the malice degree scoring S2 of this arbitrary word is less than default Evaluation threshold k3, determine that this arbitrary word is for malice word;

According to the malice word in this set of words to be identified, set up this malice word storehouse.

Optionally, this acquisition set of words to be identified, comprising:

According to the second segmentation methods preset, participle is carried out to each address in database;

Word after participle is formed this set of words to be identified.

Optionally, this second segmentation methods comprises this first participle algorithm, and this second segmentation methods comprises at least one in two character segmentation algorithms and three character segmentation algorithms.

Device 400 can also comprise the power management that a power supply module 426 is configured to actuating unit 400, and a wired or wireless network interface 450 is configured to device 400 to be connected to network, and input and output (I/O) interface 458.Device 400 can operate the operating system based on being stored in storer 432, such as WindowsServerTM, MacOSXTM, UnixTM, LinuxTM, FreeBSDTM or similar.

In sum, a kind of malice order recognition device that embodiment of the present disclosure provides, according to the first participle algorithm preset, can treat identification address and carries out participle, obtain the word to be identified in described address to be identified; And add up in described word to be identified, belong to the number of the word in malice word storehouse; According to the number of the described word belonged in described malice word storehouse, judge whether described address to be identified is malice address; When described address to be identified is malice address, determine that order corresponding to described address to be identified is for malice order.The malice order recognition methods that disclosure embodiment provides, can be identified the address in order by the malice word storehouse set up in advance, and then determines whether order is malice order, and the accuracy of this malice order recognition methods is higher.

Those skilled in the art, at consideration instructions and after putting into practice invention disclosed herein, will easily expect other embodiment of the present disclosure.The application is intended to contain any modification of the present disclosure, purposes or adaptations, and these modification, purposes or adaptations are followed general principle of the present disclosure and comprised the undocumented common practise in the art of the disclosure or conventional techniques means.Instructions and embodiment are only regarded as exemplary, and true scope of the present disclosure and spirit are pointed out by claim.

Should be understood that, the disclosure is not limited to precision architecture described above and illustrated in the accompanying drawings, and can carry out various amendment and change not departing from its scope.The scope of the present disclosure is only limited by appended claim.

Claims (11)

1. a malice order recognition methods, it is characterized in that, described method comprises:

According to the first participle algorithm preset, treat identification address and carry out participle, obtain the word to be identified in described address to be identified;

Add up in described word to be identified, belong to the number of the word in malice word storehouse, described malice word storehouse is what set up in advance, and described malice word storehouse records at least one for identifying the malice word of malice address;

According to the number of the described word belonged in described malice word storehouse, judge whether described address to be identified is malice address;

When described address to be identified is malice address, determine that order corresponding to described address to be identified is for malice order.

2. method according to claim 1, is characterized in that, belongs to the number of the word in described malice word storehouse described in described basis, judges whether described address to be identified is malice address, comprising:

Add up the number n of the word to be identified in described address to be identified;

According to the number n of the word to be identified in the number m of the described word belonged in described malice word storehouse and described address to be identified, calculated the malice degree scoring S1 of described address to be identified by address malice degree evaluate formula, described address malice degree evaluate formula is: S1=m/n;

Judge whether the malice degree scoring S1 of described address to be identified is greater than predetermined threshold value t;

When described malice degree scoring S1 is greater than predetermined threshold value t, determine that described address to be identified is for malice address;

When described malice degree scoring S1 is not more than predetermined threshold value t, determine that described address to be identified is not malice address.

3. method according to claim 1, is characterized in that, described method also comprises:

Obtain set of words to be identified;

For the arbitrary word in described set of words to be identified, calculate the malice degree scoring S2 of described arbitrary word according to word malice degree evaluate formula, described word malice degree evaluate formula is:

Wherein, abs () expression takes absolute value to content in bracket, k1 is default general character constant, k2 is default length characteristic value, tf is described arbitrary word word frequency in a database, df is described arbitrary word document frequency in the database, and L is containing the number comprising the different address of number of characters in the address of described arbitrary word in described database;

Judge whether the malice degree scoring S2 of described arbitrary word is less than default Evaluation threshold k3;

When the malice degree scoring S2 of described arbitrary word is less than default Evaluation threshold k3, determine that described arbitrary word is for malice word;

According to the malice word in described set of words to be identified, set up described malice word storehouse.

4. method according to claim 3, is characterized in that, described acquisition set of words to be identified, comprising:

According to the second segmentation methods preset, participle is carried out to each address in described database;

By the described set of words to be identified of word composition after participle.

5. method according to claim 4, is characterized in that, described second segmentation methods comprises described first participle algorithm, and described second segmentation methods comprises at least one in two character segmentation algorithms and three character segmentation algorithms.

6. a malice order recognition device, it is characterized in that, described device comprises:

Word-dividing mode, is configured to, according to the first participle algorithm preset, treat identification address and carry out participle, obtain the word to be identified in described address to be identified;

Statistical module, be configured in the described word to be identified of statistics, belong to the number of the word in malice word storehouse, described malice word storehouse is what set up in advance, and described malice word storehouse records at least one for identifying the malice word of malice address;

First judge module, is configured to the number belonging to the word in described malice word storehouse described in basis, judges whether described address to be identified is malice address;

First determination module, is configured to when described address to be identified is for malice address, determines that order corresponding to described address to be identified is for malice order.

7. device according to claim 6, is characterized in that, described first judge module, is configured to:

Add up the number n of the word to be identified in described address to be identified;

According to the number n of the word to be identified in the number m of the described word belonged in described malice word storehouse and described address to be identified, calculated the malice degree scoring S1 of described address to be identified by address malice degree evaluate formula, described address malice degree evaluate formula is: S1=m/n;

Judge whether the malice degree scoring S1 of described address to be identified is greater than predetermined threshold value t;

When described malice degree scoring S1 is greater than predetermined threshold value t, determine that described address to be identified is for malice address;

When described malice degree scoring S1 is not more than predetermined threshold value t, determine that described address to be identified is not malice address.

8. device according to claim 6, is characterized in that, described device also comprises:

Acquisition module, is configured to obtain set of words to be identified;

Computing module, is configured to for the arbitrary word in described set of words to be identified, and calculate the malice degree scoring S2 of described arbitrary word according to word malice degree evaluate formula, described word malice degree evaluate formula is: $S2=\frac{abs(tf-df)}{k1}+\frac{L}{k2};$

Wherein, abs () expression takes absolute value to content in bracket, k1 is default general character constant, k2 is default length characteristic value, tf is described arbitrary word word frequency in a database, df is described arbitrary word document frequency in the database, and L is containing the number comprising the different address of number of characters in the address of described arbitrary word in described database;

Second judge module, is configured to judge whether the malice degree scoring S2 of described arbitrary word is less than default Evaluation threshold k3;

Second determination module, is configured to when the malice degree scoring S2 of described arbitrary word is less than default Evaluation threshold k3, determines that described arbitrary word is for malice word;

Set up module, be configured to according to the malice word in described set of words to be identified, set up described malice word storehouse.

9. device according to claim 8, is characterized in that, described acquisition module, is configured to:

According to the second segmentation methods preset, participle is carried out to each address in described database;

By the described set of words to be identified of word composition after participle.

10. device according to claim 9, is characterized in that, described second segmentation methods comprises described first participle algorithm, and described second segmentation methods comprises at least one in two character segmentation algorithms and three character segmentation algorithms.

11. 1 kinds of malice order recognition devices, it is characterized in that, described device comprises:

Processor;

For storing the storer of the executable instruction of described processor;

Wherein, described processor is configured to:

According to the first participle algorithm preset, treat identification address and carry out participle, obtain the word to be identified in described address to be identified;

Add up in described word to be identified, belong to the number of the word in malice word storehouse, described malice word storehouse is what set up in advance, and described malice word storehouse records at least one for identifying the malice word of malice address;

According to the number of the described word belonged in described malice word storehouse, judge whether described address to be identified is malice address;

When described address to be identified is malice address, determine that order corresponding to described address to be identified is for malice order.