CN117715049A - Anti-cheating system and anti-cheating method for mobile phone browser - Google Patents
Anti-cheating system and anti-cheating method for mobile phone browser Download PDFInfo
- Publication number
- CN117715049A CN117715049A CN202410165498.0A CN202410165498A CN117715049A CN 117715049 A CN117715049 A CN 117715049A CN 202410165498 A CN202410165498 A CN 202410165498A CN 117715049 A CN117715049 A CN 117715049A
- Authority
- CN
- China
- Prior art keywords
- cheating
- mobile phone
- user
- potential
- users
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 230000006399 behavior Effects 0.000 claims abstract description 22
- 206010000117 Abnormal behaviour Diseases 0.000 claims abstract description 21
- 230000002159 abnormal effect Effects 0.000 claims abstract description 15
- 230000000903 blocking effect Effects 0.000 claims abstract description 14
- 230000015556 catabolic process Effects 0.000 claims abstract description 14
- 238000006731 degradation reaction Methods 0.000 claims abstract description 14
- 238000001514 detection method Methods 0.000 claims abstract description 10
- 238000003066 decision tree Methods 0.000 claims description 32
- 238000003062 neural network model Methods 0.000 claims description 11
- 238000001914 filtration Methods 0.000 claims description 9
- 238000013528 artificial neural network Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000000644 propagated effect Effects 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 5
- 230000009191 jumping Effects 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000005065 mining Methods 0.000 claims description 3
- 238000005192 partition Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 230000001902 propagating effect Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 2
- 238000007789 sealing Methods 0.000 claims description 2
- 238000012544 monitoring process Methods 0.000 abstract description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000593 degrading effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000003612 virological effect Effects 0.000 description 1
Abstract
The invention discloses an anti-cheating system and an anti-cheating method for a mobile phone browser. The method of the invention comprises the following steps: acquiring the history blocking rate, the clicking times, the IP address and the mobile phone equipment address of each mobile phone user, and identifying the first cheating mobile phone user; detecting cheating behaviors of the rest mobile phone users, judging potential cheating mobile phone users, and performing degradation operation; and extracting the characteristics of the potential cheating mobile phone users, and marking the potential cheating mobile phone users as second cheating mobile phone users if the characteristic identification is abnormal. The invention filters and degrades the users with abnormal behaviors by acquiring the user browsing logs, adopts the user behaviors, content cheating and link cheating to detect the cheating behaviors of the filtered users, and carries out matching identification on potential users by extracting features of the detected cheating users; the method has the advantages of multiple detection modes, high detection speed and wide monitoring range.
Description
Technical Field
The invention relates to the technical field of anti-cheating, in particular to an anti-cheating system and an anti-cheating method aiming at a mobile phone browser.
Background
With the rapid development of the internet, search engines have become a major approach for people to find useful information on the internet. When people find information through a search engine, only the top few search results are typically of interest. For most queries, only the first page of search results, often the first few search results, will be viewed by the searcher. The higher the ranking in the search results, the more traffic is brought to the web site. The more traffic means more profits are brought to the web site. This motivates some websites to manipulate the ranking results of the search engine by some improper method, rather than by improving their own quality. This improper way of manipulating search engine ranking results is defined as search engine cheating.
The cheating website itself has little or very low value. Search engines index them, not only wasting resources such as servers and bandwidth, but also leading to a degradation of the user experience. In addition, some fraudulent or viral websites are more self-evident from the hazards of obtaining high search engine rankings through search engine cheating. Commercial search engines have to take effective measures to reduce the negative impact of search engine cheating.
At present, the means for anti-cheating advertisements are to increase the cheating cost of cheaters, the anti-cheating purpose is achieved by making the cheating cost not proportional to the obtained benefits, and the adopted means are generally cheating detection aiming at links but have no solution for feature extraction of cheating users and extraction of potential cheating mobile phone users.
Disclosure of Invention
In order to solve the problems of single detection means and high error sealing rate of the anti-cheating system in the prior art, the invention provides the following technical scheme: a method for anti-cheating of a mobile phone browser comprises the following steps:
acquiring a browser IP, a device number and a browsing log of a mobile phone user; the equipment number information is used as the unique identification information of the user, so that the user behavior can be tracked and identified;
s1, acquiring a history blocking rate, click times, an IP address and a mobile phone equipment address of each mobile phone user;
s2, identifying the history blocking rate, the clicking times, the IP address and the mobile phone equipment address of each mobile phone user, judging whether the mobile phone user is a cheating mobile phone user, marking the cheating mobile phone user obtained through identification as a first cheating mobile phone user, and adding the first cheating mobile phone user into a blacklist;
s3, filtering the first cheating mobile phone user, and enabling the rest mobile phone users to enter the next step of processing;
s4, detecting cheating behaviors of each rest mobile phone user, judging the mobile phone user as a potential cheating mobile phone user if the cheating behaviors occur, and performing degradation operation on the potential cheating mobile phone user;
meanwhile, extracting the characteristics of the potential cheating mobile phone user, carrying out characteristic identification through a neural network algorithm, and marking the potential cheating mobile phone user as a second cheating mobile phone user if the characteristic identification is abnormal;
s5, adding the second cheating mobile phone user to the blacklist.
Preferably, the first cheating mobile phone user is defined as having one or more of the following actions: high blocking rate, abnormal click times, and different IP addresses between IP positioning and mobile phone real-time positioning
Preferably, each remaining mobile phone user after filtering is subjected to cheating detection, and one or more of the following behaviors exist, wherein the behaviors are defined as abnormal behaviors:
detecting user behaviors, namely detecting abnormal behavior characteristics of a user in a browsing log; the method comprises the following steps:
(1) The user browses all web page sources in the log, wherein the click rate of the search engine is too high;
(2) In the browsing log, in all records containing a certain webpage, the proportion of the user jumping from the webpage to other webpages is abnormal;
(3) In the browsing log, all the search words contained in a certain webpage record have a great number of repeated abnormal behaviors with higher frequency;
(4) The browsing or clicking actions are generated in a plurality of advertisement positions by the same user at the same time, or the same advertisement position is subjected to multiple exposure or clicking in a short time;
if abnormal behaviors are detected, judging the users with the abnormal behaviors as potential cheating mobile phone users, and performing degradation operation on the potential cheating mobile phone users;
detecting tf and idf values of the rest mobile phone users and calculating the relevance of the searched web pages and search words; the cheating of the webpage content usually repeatedly occurs a large number of keywords set by cheaters and content irrelevant to search words in the webpage, and whether the webpage is the cheating webpage is judged by calculating tf and idf values of the browser and according to the relevance of the searched webpage and the search words; the method comprises the following steps:
where t represents a particular word, d represents the content in the web page, f (t) represents the number of times the particular word appears in the web page,representing the number of times all words appear in the web page; the frequency of occurrence of a specific word can be obtained through calculation;
where n represents the total number of web page contents, df (t) is the number of web pages containing a specific word t,；
further, the frequency of occurrence of the specific word in the searched web page content can be obtained through the calculated tf (t, d) and idf (t);
further, a plurality of keywords or query strings are generally searched during searching, and the relevance between the keywords or the query strings and the webpage content is calculated; the method comprises the following steps:
in the method, in the process of the invention,representing a web page content vector, ">Query vector for user->Modulo the web page content vector, +.>A module representing a user query vector; if the calculated value is smaller, the correlation between the webpage content and the query string is higher, and if the calculated value is higher, the correlation is lower;
if the tf and idf values of the search content and the web page content are higher, but the correlation is lower, the web page is judged to be cheated, and the degradation operation is carried out on the mobile phone user judged to be potential cheating by the user.
Preferably, degradation to detected potentially cheating handset users:
selecting a set containing high-quality webpages as a seed set;
recursively propagating trust to the entire web, starting from the seed set, along the direction of the links;
after convergence, each propagated webpage can obtain a trust score, and the trust score of the propagated webpage is calculated; the calculation formula is as follows:
wherein g represents trust score, alpha is attenuation factor, M is adjacency matrix, s is seed set trust score vector;
the lower the trust score, the later the web pages will rank in the search engine when the user searches next.
Preferably, the features of the potentially cheating mobile phone user are extracted:
extracting the characteristics of the potential cheating mobile phone user by using a decision tree algorithm according to the detected cheating webpage data; the method comprises the following steps:
acquiring a user browsing log, and extracting characteristics of a cheating user; obtaining a training set by randomly sampling cheating users and constructing a decision tree model; then the cheating features of the potential users are mined through the decision tree model;
the cheating features of mining potential users are as follows:
inputting the potential user data into a decision tree model, comparing the information gain of each node in the potential user data and the decision tree model, and selecting the next branch according to the decision result until the decision result is finally obtained as the potential user cheating feature;
the decision result is a comparison result of the information gain of each node in the potential user data and the decision tree model;
the decision tree algorithm model is constructed specifically as follows:
(1) Randomly sampling samples, and selecting N samples to train a decision tree;
(2) Starting from the root node, calculating information gain of all features in the data set corresponding to the node, and selecting the feature with the maximum information gain value as the dividing feature of the data set;
(3) Establishing subtrees for the corresponding data set partitions of the feature;
(4) Recursively calling the step (2) and the step (3) for the sub-tree to construct a decision tree;
(5) Until the information gain of the data set feature reaches a certain condition or all samples belong to the same type, the final decision tree is obtained.
Preferably, the extracted potentially cheating mobile phone user features are identified by a neural network algorithm:
the data of the training set is normalized and then is input into a neural network model, and the result tends to converge through continuous forward propagation and error reverse propagation, so that a trained neural network model is obtained;
the extracted potential user cheating features are used as cheating features to be input into a trained neural network model, and the cheating features are identified with the cheating features in the database;
if the feature identification is abnormal, the next step is carried out.
The invention also discloses a system for preventing cheating aiming at the mobile phone browser, which realizes the method:
the filtering module is used for identifying the history blocking rate, the clicking times, the IP address and the mobile phone equipment address of each mobile phone user;
the detection module is used for detecting cheating behaviors of each rest mobile phone user, judging the mobile phone user as a potential cheating mobile phone user if the cheating behaviors occur, and performing degradation operation on the potential cheating mobile phone user;
the identification module is used for extracting the characteristics of the potential cheating mobile phone user and carrying out characteristic identification through a neural network algorithm;
and the blacklist module is used for establishing a blacklist.
Compared with the prior art, the invention provides a system and a method for anti-cheating aiming at the mobile phone browser, which have the following beneficial effects:
according to the method, the mobile phone equipment number information is obtained and used as the unique identification information, so that the influence of the virtual machine can be avoided.
According to the method, the historical blocking rate, the clicking times, the IP address and the mobile phone equipment address of the user are obtained through obtaining the mobile phone browsing log, the user with abnormal data is filtered and added into the blacklist, and anti-cheating pressure is reduced.
The invention detects various cheating modes and has wider monitoring degree.
The invention improves the anti-cheating efficiency by extracting the characteristics of the cheating user by using the decision tree and exploring the cheating user by using the neural network model.
Drawings
FIG. 1 is a flow chart of a system and method for anti-cheating in a mobile phone browser according to the present invention;
FIG. 2 is a schematic diagram of a system and method for anti-cheating in a mobile phone browser according to the present invention;
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1: s1, acquiring a mobile phone user browser IP, a device number and a browsing log;
further, the device number information is used as the unique identification information of the user;
s2, identifying the history blocking rate, the clicking times, the IP address and the mobile phone equipment address of each mobile phone user, judging whether the mobile phone user is a cheating mobile phone user, marking the cheating mobile phone user obtained through identification as a first cheating mobile phone user, and adding the first cheating mobile phone user into a blacklist;
the first cheating handset user is defined as having one or more of the following actions: high blocking rate, abnormal click times, and different IP addresses between IP positioning and mobile phone real-time positioning; wherein, the blocking rate is higher than 60% and is judged as high, and the clicking times are higher than 100 times per minute and the clicking times of the judging site are abnormal
S3, filtering the first cheating mobile phone user, and performing cheating detection on the filtered user;
detecting user behaviors, namely detecting abnormal behavior characteristics of a user in a browsing log; there is one or more of the following behaviors, defined as abnormal behavior:
(1) The user browses all web page sources in the log, wherein the click rate of the search engine is too high, and further, the too high is defined as more than 70%;
(2) In the browsing log, in all records containing a certain webpage, the proportion of the user jumping from the webpage to other webpages is abnormal, and the proportion of the user jumping from the webpage to other webpages reaches 75% or more, so that the abnormal behavior is realized;
(3) In the browsing log, all the search words contained in a certain webpage record have a great number of repeated abnormal behaviors with higher frequency; further, a frequency higher than 100 times per minute is defined as abnormal behavior;
(4) The browsing or clicking actions are generated in a plurality of advertisement positions by the same user at the same time, or the same advertisement position is subjected to multiple exposure or clicking in a short time; further, the behavior of generating 50 exposures or clicks per minute in the same advertisement spot is abnormal behavior;
if abnormal behaviors are detected, judging the users with the abnormal behaviors as potential cheating mobile phone users, and performing degradation operation on the potential cheating mobile phone users.
Further, detecting tf and idf values of each rest of mobile phone users and calculating the relevance between the searched web pages and the search words; the cheating of the webpage content usually repeatedly occurs a large number of keywords set by cheaters and content irrelevant to search words in the webpage, and whether the webpage is the cheating webpage is judged by calculating tf and idf values of the browser and according to the relevance of the searched webpage and the search words; the method comprises the following steps:
where t represents a particular word, d represents the content in the web page, f (t) represents the number of times the particular word appears in the web page,representing all words in a web pageThe number of occurrences; the frequency of occurrence of a specific word can be obtained through calculation;
where n represents the total number of web page contents, df (t) is the number of web pages containing a specific word t,；
further, the frequency of occurrence of the specific word in the searched web page content can be obtained through the calculated tf (t, d) and idf (t);
further, a plurality of keywords or query strings are generally searched during searching, and the relevance between the keywords or the query strings and the webpage content is calculated; the method comprises the following steps:
in the method, in the process of the invention,representing a web page content vector, ">Query vector for user->Modulo the web page content vector, +.>A module representing a user query vector; if the calculated value is smaller, the correlation between the webpage content and the query string is higher, and if the calculated value is higher, the correlation is lower;
if the tf and idf values of the search content and the web page content are higher, but the correlation is lower, judging that the web page is cheating, if abnormal behaviors are detected, judging that the user with the abnormal behaviors is a potential cheating mobile phone user, and performing degradation operation on the potential cheating mobile phone user;
meanwhile, degrading the detected potential cheating mobile phone user, and reducing the priority of the potential cheating mobile phone user;
selecting a set containing high-quality webpages as a seed set;
recursively propagating trust to the entire web, starting from the seed set, along the direction of the links;
after convergence, each propagated webpage can obtain a trust score, and the trust score of the propagated webpage is calculated; the calculation formula is as follows:
wherein g represents trust score, alpha is attenuation factor, M is adjacency matrix, s is seed set trust score vector;
the lower the trust score, the later the web pages will be ranked in the search engine when the user searches next time;
s4, extracting the characteristics of the potential cheating mobile phone user, and identifying through a neural network algorithm;
extracting cheating user features by using a decision tree algorithm according to the detected cheating webpage data; the method comprises the following steps:
acquiring a user browsing log, and extracting characteristics of a cheating user; obtaining a training set by randomly sampling cheating users and constructing a decision tree model; then the cheating features of the potential users are mined through the decision tree model;
the cheating features of mining potential users are as follows:
inputting the potential user data into a decision tree model, comparing the information gain of each node in the potential user data and the decision tree model, and selecting the next branch according to the decision result until the decision result is finally obtained as the potential user cheating feature;
the decision result is a comparison result of the information gain of each node in the potential user data and the decision tree model;
the decision tree algorithm model is constructed specifically as follows:
(1) The samples are randomly sampled, N samples are selected to train a decision tree, and the input of all data can be avoided to a certain extent, so that the overfitting is avoided;
(2) Starting from the root node, calculating information gain of all features in the data set corresponding to the node, and selecting the feature with the maximum information gain value as the dividing feature of the data set;
(3) Establishing subtrees for the corresponding data set partitions of the feature;
(4) Recursively calling the step (2) and the step (3) for the sub-tree to construct a decision tree;
(5) Until the information gain of the data set characteristics reaches a certain condition or all samples belong to the same type, obtaining a final decision tree;
further, the extracted potential cheating mobile phone user features are used as cheating features to be input into a trained neural network model, and similarity matching is carried out on the cheating features and the cheating features in the database; if the similarity reaches 80% or more, if the feature is abnormal, the next step is carried out.
The data of the training set is normalized and then is input into a neural network model, and the result is converged through continuous forward propagation and error reverse propagation, so that a trained neural network model is obtained;
if the matching is successful, determining that the potential user has cheating behaviors, wherein the potential cheating mobile phone user is marked as a second cheating mobile phone user;
s5, the second cheating mobile phone user is added to the blacklist.
Example 2: the embodiment discloses a system for anti-cheating of a mobile phone browser, which comprises the following modules:
the filtering module is used for detecting the history blocking rate, the clicking times, the IP addresses and the mobile phone equipment addresses in the browsing log, filtering the IP addresses with high blocking rate, abnormal clicking times and different IP positioning from the mobile phone real-time positioning and recording the IP addresses in a blacklist;
the detection module is used for detecting the search content and the link cheating through the user behavior and carrying out degradation operation on the user;
the identification module is used for extracting the characteristics of the cheating user by using a decision tree algorithm, and inputting the characteristics of the cheating user into the neural network model for matching identification;
and the blacklist module is used for establishing a blacklist, and adding the filtered cheating users and the checked cheating users into the blacklist to limit the search performance of the browser.
While embodiments of the present invention have been shown and described in use, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (8)
1. The anti-cheating method for the mobile phone browser is characterized by comprising the following steps of:
s1, acquiring a history blocking rate, click times, an IP address and a mobile phone equipment address of each mobile phone user;
s2, identifying the history blocking rate, the clicking times, the IP address and the mobile phone equipment address of each mobile phone user, judging whether the mobile phone user is a cheating mobile phone user, marking the cheating user obtained by identification as a first cheating mobile phone user, and adding the first cheating mobile phone user into a blacklist;
s3, filtering the first cheating mobile phone user, and enabling the rest mobile phone users to enter the next step of processing;
s4, detecting cheating behaviors of each rest mobile phone user, judging the mobile phone user as a potential cheating mobile phone user if the cheating behaviors occur, and performing degradation operation on the potential cheating mobile phone user;
meanwhile, extracting the characteristics of the potential cheating mobile phone user, carrying out characteristic identification through a neural network algorithm, and marking the potential cheating mobile phone user as a second cheating mobile phone user if the characteristic identification is abnormal;
s5, adding the second cheating mobile phone user to the blacklist.
2. The method for preventing cheating in a mobile phone browser according to claim 1, wherein,
the first cheating handset user is defined as having one or more of the following actions: the sealing rate is high, the clicking times are abnormal, and the IP addresses with different IP positioning and mobile phone real-time positioning are realized.
3. The anti-cheating method for a mobile phone browser according to claim 1, wherein the cheating detection is performed on each remaining mobile phone user after filtering:
detecting user behaviors, namely detecting abnormal behavior characteristics of a user in a browsing log; there is one or more of the following behaviors, defined as abnormal behavior:
(1) The user browses all web page sources in the log, wherein the click rate of the search engine is too high;
(2) In the browsing log, in all records containing a certain webpage, the proportion of the user jumping from the webpage to other webpages is abnormal;
(3) In the browsing log, all the search words contained in a certain webpage record have a great number of repeated abnormal behaviors with higher frequency;
(4) The browsing or clicking actions are generated in a plurality of advertisement positions by the same user at the same time, or the same advertisement position is subjected to multiple exposure or clicking in a short time;
if abnormal behaviors are detected, judging the users with the abnormal behaviors as potential cheating mobile phone users, and performing degradation operation on the potential cheating mobile phone users.
4. The method for preventing cheating in a mobile phone browser according to claim 1, wherein,
detecting tf and idf values of the rest mobile phone users and calculating the relevance of the searched web pages and search words; judging whether the web page is a cheating web page or not according to the relevance of the searched web page and the search word by calculating tf and idf values of the browser; the method comprises the following steps:
；
where t represents a particular word, d represents the content in the web page, f (t) represents the number of times the particular word appears in the web page,representing the number of times all words appear in the web page; the frequency of occurrence of a specific word can be obtained through calculation;
；
where n represents the total number of web page contents, df (t) is the number of web pages containing a specific word t,；
the frequency of occurrence of the specific word in the searched web page content can be obtained through the calculated tf (t, d) and idf (t);
searching a plurality of keywords or query strings during searching, and calculating the relativity between the keywords or query strings and the webpage content; the method comprises the following steps:
；
in the method, in the process of the invention,representing a web page content vector, ">Query vector for user->Modulo the web page content vector, +.>Representing user searchesModulo of the polling vector; if the calculated value is smaller, the correlation between the webpage content and the query string is higher, and if the calculated value is higher, the correlation is lower;
if tf and idf values of the search content and the web page content are higher, but the correlation is lower, judging that the web page is cheated, and performing degradation operation on the user judged that the user is a potential cheating mobile phone user.
5. The anti-cheating method for a mobile phone browser of claim 1, wherein the detected potential cheating mobile phone user is downgraded:
selecting a set containing high-quality webpages as a seed set;
recursively propagating trust to the entire web, starting from the seed set, along the direction of the links;
after convergence, each propagated webpage can obtain a trust score, and the trust score of the propagated webpage is calculated; the calculation formula is as follows:
；
wherein g represents trust score, alpha is attenuation factor, M is adjacency matrix, s is seed set trust score vector;
the lower the trust score, the later the web pages will rank in the search engine when the user searches next.
6. The anti-cheating method for a mobile phone browser according to claim 1, wherein features of potential cheating mobile phone users are extracted:
extracting the characteristics of the potential cheating mobile phone user by using a decision tree algorithm according to the detected cheating webpage data; the method comprises the following steps:
acquiring a user browsing log, and extracting characteristics of a cheating user; obtaining a training set by randomly sampling cheating users and constructing a decision tree model; then the cheating features of the potential users are mined through the decision tree model;
the cheating features of mining potential users are as follows:
inputting the potential user data into a decision tree model, comparing the information gain of each node in the potential user data and the decision tree model, and selecting the next branch according to the decision result until the decision result is finally obtained as the potential user cheating feature;
the decision result is a comparison result of the information gain of each node in the potential user data and the decision tree model;
the decision tree algorithm model is constructed specifically as follows:
(1) Randomly sampling samples, and selecting N samples to train a decision tree;
(2) Starting from the root node, calculating information gain of all features in the data set corresponding to the node, and selecting the feature with the maximum information gain value as the dividing feature of the data set;
(3) Establishing subtrees for the corresponding data set partitions of the feature;
(4) Recursively calling the step (2) and the step (3) for the sub-tree to construct a decision tree;
(5) Until the information gain of the data set feature reaches a certain condition or all samples belong to the same type, the final decision tree is obtained.
7. The anti-cheating method for a mobile phone browser according to claim 1, wherein the extracted potential cheating mobile phone user features are identified by a neural network algorithm:
the data of the training set is normalized and then is input into a neural network model, and the result tends to converge through continuous forward propagation and error reverse propagation, so that a trained neural network model is obtained;
the extracted potential user cheating features are used as cheating features to be input into a trained neural network model, and the cheating features are identified with the cheating features in the database;
if the feature identification is abnormal, the next step is carried out.
8. A mobile browser anti-cheating system for implementing the anti-cheating method for a mobile browser of any one of claims 1-7, comprising the following modules:
the filtering module is used for judging whether each mobile phone user is a cheating mobile phone user or not by identifying the historical blocking rate, the clicking times, the IP address and the mobile phone equipment address of the mobile phone user;
the detection module is used for detecting cheating behaviors of each rest mobile phone user, judging the mobile phone user as a potential cheating mobile phone user if the cheating behaviors occur, and performing degradation operation on the potential cheating mobile phone user;
the identification module is used for extracting the characteristics of the potential cheating mobile phone user and carrying out characteristic identification through a neural network algorithm;
and the blacklist module is used for establishing a blacklist.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410165498.0A CN117715049B (en) | 2024-02-05 | 2024-02-05 | Anti-cheating system and anti-cheating method for mobile phone browser |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410165498.0A CN117715049B (en) | 2024-02-05 | 2024-02-05 | Anti-cheating system and anti-cheating method for mobile phone browser |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117715049A true CN117715049A (en) | 2024-03-15 |
| CN117715049B CN117715049B (en) | 2024-04-12 |
Family
ID=90159297
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410165498.0A Active CN117715049B (en) | 2024-02-05 | 2024-02-05 | Anti-cheating system and anti-cheating method for mobile phone browser |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117715049B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102243659A (en) * | 2011-07-18 | 2011-11-16 | 南京邮电大学 | Webpage junk detection method based on dynamic Bayesian model |
| CN102254265A (en) * | 2010-05-18 | 2011-11-23 | 北京首家通信技术有限公司 | Rich media internet advertisement content matching and effect evaluation method |
| CN102289756A (en) * | 2010-06-18 | 2011-12-21 | 百度在线网络技术（北京）有限公司 | Method and system for judging click validation |
| US20150262069A1 (en) * | 2014-03-11 | 2015-09-17 | Delvv, Inc. | Automatic topic and interest based content recommendation system for mobile devices |
| CN105677869A (en) * | 2016-01-06 | 2016-06-15 | 广州神马移动信息科技有限公司 | Multidimensional search log anti-cheating method, system and computing equipment |
| US20160203222A1 (en) * | 2013-08-21 | 2016-07-14 | Baidu Online Network Technology (Beijing) Co., Ltd. | Search method, search system, and search engine |
| US20200155944A1 (en) * | 2018-11-19 | 2020-05-21 | Nant Holdings Ip, Llc | Distributed ledger tracking of event data |
| CN111882367A (en) * | 2020-08-11 | 2020-11-03 | 北京云和互动信息技术有限公司 | Method for monitoring and tracking online advertisements through user internet behavior analysis |
| CN112734472A (en) * | 2021-01-12 | 2021-04-30 | 上海浩御文化传媒有限公司 | Advertisement traffic anti-cheating method |
| CN113791837A (en) * | 2021-08-12 | 2021-12-14 | 百度在线网络技术（北京）有限公司 | Page processing method, device, equipment and storage medium |
| CN116797282A (en) * | 2023-08-28 | 2023-09-22 | 成都一心航科技有限公司 | Real-time monitoring system and monitoring method for advertisement delivery |
-
2024
- 2024-02-05 CN CN202410165498.0A patent/CN117715049B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102254265A (en) * | 2010-05-18 | 2011-11-23 | 北京首家通信技术有限公司 | Rich media internet advertisement content matching and effect evaluation method |
| CN102289756A (en) * | 2010-06-18 | 2011-12-21 | 百度在线网络技术（北京）有限公司 | Method and system for judging click validation |
| CN102243659A (en) * | 2011-07-18 | 2011-11-16 | 南京邮电大学 | Webpage junk detection method based on dynamic Bayesian model |
| US20160203222A1 (en) * | 2013-08-21 | 2016-07-14 | Baidu Online Network Technology (Beijing) Co., Ltd. | Search method, search system, and search engine |
| US20150262069A1 (en) * | 2014-03-11 | 2015-09-17 | Delvv, Inc. | Automatic topic and interest based content recommendation system for mobile devices |
| CN105677869A (en) * | 2016-01-06 | 2016-06-15 | 广州神马移动信息科技有限公司 | Multidimensional search log anti-cheating method, system and computing equipment |
| US20200155944A1 (en) * | 2018-11-19 | 2020-05-21 | Nant Holdings Ip, Llc | Distributed ledger tracking of event data |
| CN111882367A (en) * | 2020-08-11 | 2020-11-03 | 北京云和互动信息技术有限公司 | Method for monitoring and tracking online advertisements through user internet behavior analysis |
| CN112734472A (en) * | 2021-01-12 | 2021-04-30 | 上海浩御文化传媒有限公司 | Advertisement traffic anti-cheating method |
| CN113791837A (en) * | 2021-08-12 | 2021-12-14 | 百度在线网络技术（北京）有限公司 | Page processing method, device, equipment and storage medium |
| CN116797282A (en) * | 2023-08-28 | 2023-09-22 | 成都一心航科技有限公司 | Real-time monitoring system and monitoring method for advertisement delivery |
Non-Patent Citations (3)
| Title |
|---|
| ALBERTO PASTOR NIETO ECT.: "How Android protects you from scams and phishing attacks", Retrieved from the Internet <URL:https://blog.google/products/android/android-scam-and-phishing-protection/> * |
| 余慧佳;刘奕群;张敏;马少平;茹立云;: "基于目的分析的作弊页面分类", 中文信息学报, no. 02, 15 March 2009 (2009-03-15) * |
| 李智超;余慧佳;刘奕群;马少平;: "网页作弊与反作弊技术综述", 山东大学学报(理学版), no. 05, 4 May 2011 (2011-05-04) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117715049B (en) | 2024-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10482136B2 (en) | Method and apparatus for extracting topic sentences of webpages | |
| US7693904B2 (en) | Method and system for determining relation between search terms in the internet search system | |
| CN103559235B (en) | A kind of online social networks malicious web pages detection recognition methods | |
| US8244752B2 (en) | Classifying search query traffic | |
| US9443019B2 (en) | Optimized web domains classification based on progressive crawling with clustering | |
| US8335785B2 (en) | Ranking results for network search query | |
| US7636714B1 (en) | Determining query term synonyms within query context | |
| US6970863B2 (en) | Front-end weight factor search criteria | |
| US8219549B2 (en) | Forum mining for suspicious link spam sites detection | |
| US20100023508A1 (en) | Search engine enhancement using mined implicit links | |
| US20110208714A1 (en) | Large scale search bot detection | |
| EP1653380A1 (en) | Web page ranking with hierarchical considerations | |
| JP2008181186A (en) | Method for determining relevancy between keyword and web site using query log | |
| CN101093510A (en) | Anti cheating method and system for aiming at cheat on web page | |
| JP2009503751A (en) | Relevance determination method, information collection method, object organization method, and search system | |
| WO2009000174A1 (en) | Method and device of web page rank | |
| CN107798080B (en) | Similar sample set construction method for fishing URL detection | |
| CN109104421B (en) | Website content tampering detection method, device, equipment and readable storage medium | |
| CN111324801B (en) | Hot event discovery method in judicial field based on hot words | |
| US10467255B2 (en) | Methods and systems for analyzing reading logs and documents thereof | |
| CN113297457B (en) | High-precision intelligent information resource pushing system and pushing method | |
| CN114330329A (en) | Service content searching method and device, electronic equipment and storage medium | |
| CN114048389B (en) | Content recommendation method and system for engineering machinery industry | |
| CN105389328A (en) | Method for optimizing search sorting of large-scale open source software | |
| Choudhary et al. | Role of ranking algorithms for information retrieval |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |