WO2012145962A1 - 彩铃web系统防止恶意订购铃音的方法及服务器 - Google Patents

彩铃web系统防止恶意订购铃音的方法及服务器 Download PDF

Info

Publication number
WO2012145962A1
WO2012145962A1 PCT/CN2011/076205 CN2011076205W WO2012145962A1 WO 2012145962 A1 WO2012145962 A1 WO 2012145962A1 CN 2011076205 W CN2011076205 W CN 2011076205W WO 2012145962 A1 WO2012145962 A1 WO 2012145962A1
Authority
WO
WIPO (PCT)
Prior art keywords
user
request
information
server
login
Prior art date
Application number
PCT/CN2011/076205
Other languages
English (en)
French (fr)
Inventor
令周扬
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2012145962A1 publication Critical patent/WO2012145962A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L63/00—Network architectures or network communication protocols for network security
    • H04L63/14—Network architectures or network communication protocols for network security for detecting or protecting against malicious traffic
    • H04L63/1441—Countermeasures against malicious traffic
    • H04L63/1483—Countermeasures against malicious traffic service impersonation, e.g. phishing, pharming or web spoofing

Definitions

  • the present invention relates to the field of communications technologies, and in particular, to a browser/Severage (B/S) architecture color ring back tone web system for preventing malicious ordering of ring tones and a monthly server.
  • B/S browser/Severage
  • the CRBT service is applied more and more widely.
  • the web server-based color ring tone display, audition, ordering and gifting are all carried out through the Internet.
  • the openness of the network makes the security and effectiveness of the network interaction on this basis very important.
  • the main object of the present invention is to provide a method and a server for preventing a malicious ringing tone from being generated by a CRBT Web system, so as to improve the security of the CRBT Web system.
  • the present invention provides a method for preventing a malicious ringing tone by using a CRBT Web system, including:
  • the server records the active path information of the user's mandatory step in the process of ordering the ring tone, and pushes the active path information into the stack sensitive information; Receiving a subscription request input by the user, and pressing the subscription request into the stack sensitive information; and calculating, according to the stack sensitive information, a credibility indicator of the subscription request by using a predetermined association iteration algorithm;
  • the order request is filtered out.
  • the mandatory step comprises: login, search and/or browsing.
  • the activity path information of the login process includes a page verification code component, a request IP address, a user number, and/or current request page address information.
  • the method further comprises: allowing the subscription request to pass when the credibility indicator does not exceed a predetermined threshold.
  • the method further comprises:
  • the server When receiving the login request sent by the user, the server generates a token field according to the login request, and carries the token field in the login response message and returns it to the user.
  • the method further comprises: when receiving the browsing or search request sent by the user, the server updates the token field according to the browsing or search request, and updates The subsequent token field is carried back to the user in a browse or search response message.
  • the method further comprises:
  • the server When receiving a request for querying the specified ringtone information sent by the user, the server updates the token field according to the request for the query to specify the ringtone information, and carries the updated token field in the query for the specified ringtone information.
  • the response message is returned to the user.
  • the present invention further provides a server for preventing a malicious ringing tone by a CRBT Web system, including: a stack module, configured to record active path information of a user's mandatory step in ordering a ring tone, and push the active path information into a stack Sensitive information;
  • a receiving module configured to receive a subscription request input by a user, and press the subscription request into the location Stack sensitive information
  • a calculation module configured to calculate a credibility indicator of the subscription request by using a predetermined association iteration algorithm based on the stack sensitive information
  • the order request processing module is configured to filter out the order request when the credibility indicator exceeds a predetermined threshold.
  • the order request processing module is further configured to allow the order request to pass when the credibility indicator does not exceed a predetermined threshold.
  • the method further comprises:
  • the token module is configured to generate a token field according to the login request when the server receives the login request sent by the user, and carry the token field in the login response message and return it to the user.
  • the token module is further configured to: when the server receives the browsing or search request sent by the user, update the token field according to the browsing or search request, and carry the updated token field Returned to the user in a browse or search response message.
  • the token module is further configured to, when the server receives the request for querying the specified ringtone information sent by the user, update the token field according to the request for querying the ringtone information, and update the token field.
  • the token field is returned to the user in a response message for querying the specified ring tone information.
  • the method and server for preventing a malicious ring tone from being processed by the CRBT Web system by introducing a stack mechanism, recording the activity path information of a mandatory step of the user in the process of ordering the ring tone, and pushing the activity path information into Stack sensitive information; based on the stack sensitive information, using a predetermined association iteration algorithm to calculate a credibility indicator of the subscription request; when the credibility indicator exceeds a predetermined threshold, filtering out the subscription request, thereby Security is effectively controlled; at the same time, based on this, a token-like ring mechanism is further introduced, and a token field generated according to the request of the user mandatory step is carried in the corresponding response message and returned to the user, and according to the subsequent
  • the user's request for association update of the token field can largely prevent the normal message from being sent after repeated transmission or tampering after being captured by the hacker.
  • security control is implemented on the way that the program simulating robot impersonates a legitimate user to log in, accesses through a browser address, legally logs in a user, and multiple malicious submissions, thereby improving the security of the Web system.
  • the common interests between CRBT users, operators and CRBT content providers are guaranteed.
  • FIG. 1 is a flow chart showing an embodiment of a method for preventing a malicious ringing tone from being silenced by a CRBT Web system according to the present invention
  • FIG. 2 is a schematic flowchart of a method for the server to record the active path information of a mandatory step in the process of ordering a ring tone, and to push the active path information into the stack sensitive information in an embodiment of the method for preventing the malicious ringing of the ring tones in the color ring network system of the present invention
  • FIG. 3 is a schematic diagram of a flow chart of a method for the server to record the mandatory path of the user in the process of ordering the ring tone, and the active path information is pushed into the stack sensitive information in another embodiment of the method for preventing the malicious ringing of the ring tones in the CRBT web system of the present invention.
  • FIG. 4 is a schematic structural diagram of an embodiment of a server for preventing a malicious subscription of a ring tone by the CRBT Web system of the present invention
  • FIG. 5 is a block diagram showing another embodiment of a server for preventing a malicious order ring tone from the CRBT Web system of the present invention. detailed description
  • the main idea of the solution of the embodiment of the present invention is: Since the ringtone malicious ordering behavior mostly presents the following main features: behavioral anomalies, such as behaviors that ordinary users have little or no difficulty to do; behavioral regularity, such as the same Or a sequence of operating time periods; regularity on the data, such as the natural sequence relationship of all ordered ringtone numbers, etc., therefore, Embodiments are directed to the above situation, a path tracking technology introduced by a stack mechanism for a malicious ring tone ordering problem caused by a program simulating a robot impersonating a legitimate user to log in; a malicious bell for a browser address access and a legal login user through multiple submissions Tone subscription, introduces a request filter of the Token Ring mechanism to effectively control the security of the CRBT Web system.
  • behavioral anomalies such as behaviors that ordinary users have little or no difficulty to do
  • behavioral regularity such as the same Or a sequence of operating time periods
  • regularity on the data such as the natural sequence relationship of all ordered ringtone numbers, etc.
  • the server in the following embodiment is a web server.
  • an embodiment of the present invention provides a method for preventing a malicious ringing tone by using a CRBT Web system, including:
  • Step S101 The server records the active path information of the user's mandatory step in the process of ordering the ring tone, and pushes the active path information into the stack sensitive information.
  • a malicious ring tone ordering problem caused by a program impulsively impersonating a legitimate user to log in is introduced, and a path tracking technology of the stack mechanism is introduced to dynamically track the activity state of the user on the CRBT web system, and the status of the state change is determined. Whether the current ordering operation is initiated by the user, the abnormal state behavior is considered an invalid ordering operation.
  • the ordinary CRBT user can be simplified from the login to the subscription behavior as the following key operations: login, browse/search, display, initiate a subscription based on the ring tone number, complete the order and prompt.
  • illegal users only include some key operations such as ordering.
  • illegal users who are maliciously ordered usually use the Uniform Resource Locator (URL) address shortcut, or intercept and tamper with legitimate messages, and repeat the transmission to complete the order quickly.
  • URL Uniform Resource Locator
  • some mandatory steps such as login, search, browsing, etc. will not be executed, and these operations that initiate malicious ordering are performed very frequently, and the request parameters also conform to certain rules, because The implementation of the software does not really simulate the actual random request.
  • the ringtone that is finally ordered may belong to only one SP (Service Provider).
  • this embodiment records technical steps such as a user requesting channel, such as adding a page verification code component, requesting an IP address, a user number, a current request page address, and the like in the login request, and recording the necessary steps of the user at the Web site.
  • Activity path and push this information into The stack information is popped up during the last ordering operation, and the algorithm of the class association rule is used to calculate the credibility indicator of the order request.
  • Step S102 receiving a subscription request input by the user, and pushing the subscription request into the stack sensitive information.
  • Step S103 based on the stack sensitive information, calculate a credibility indicator of the subscription request by using a predetermined association iteration algorithm.
  • Step S104 When the credibility indicator exceeds a predetermined threshold, the order request is filtered out.
  • the legality of the user's order request is judged by the credibility indicator of the order request.
  • the credibility indicator of the order request is obtained through an association iterative algorithm.
  • the calculation formula of the association iterative algorithm is:
  • mandatory step x-1) refers to the mandatory step x-1
  • the conditional probability of the mandatory step X is executed.
  • the corresponding active path information needs to be extracted from the stack sensitive information to obtain a mandatory step for the user to log in to the ordering process.
  • conditional probability is the conditional probability that time A occurs for any event A and event B under the condition that the known time B occurs.
  • the formula for calculating the conditional probability is:
  • P ( AB ) is the number of sample points contained in A under the condition of B
  • P ( B ) is the number of sample points under the condition of B occurrence.
  • the current operation is that the reliability of the normal user depends on the credibility of all the necessary steps.
  • the path length increases and whether there is a jump between the page and the page, etc.
  • the credibility index the more the normal user, the higher the credibility index, and vice versa.
  • the abnormal order request can be filtered out by a preset threshold.
  • the credibility indicator does not exceed the predetermined threshold, indicating that the subscriber is a normal subscriber, the subscription request is allowed to pass.
  • step S101 specifically includes:
  • Step S1011 Receive a login request sent by the user, record a page verification code component in the login request, request an IP address, a user number, and/or an active path information of a current request page address, and push the active path information of the login request onto the stack. Sensitive information;
  • Step S1012 Return a login request response message to the user.
  • Step S1013 Receive a browsing/search request sent by the user, and press the activity path information of the browsing/search request into the stack sensitive information;
  • Step S1014 returning a browsing/search request response message to the user
  • Step S1015 Receive a request for a specified ring tone information sent by the user, and press the active path information of the specified ring tone information request into the stack sensitive information.
  • Step S1016 Return a specified ring tone information request response message to the user.
  • the active path information of the mandatory step of the Web site in the process of ordering the ring tone is recorded, and the active path information is pushed into the stack sensitive information; based on the stack sensitive information, a predetermined association iteration is used.
  • the algorithm calculates a credibility indicator of the subscription request; when the credibility indicator exceeds a predetermined threshold, the subscription request is filtered out, thereby the CRBT web system.
  • step S101 specifically includes:
  • Step S1011 Receive a login request sent by the user, record a page verification code component in the login request, request an IP address, a user number, and/or an active path information of a current request page address, and push the active path information of the login request onto the stack. Sensitive information;
  • Step S10112 Generate a token field according to the login request, and carry the token field in the login response message.
  • Step S1012 Return a login request response message to the user.
  • Step S1013 Receive a browsing/search request sent by the user, and press the activity path information of the browsing/search request into the stack sensitive information;
  • Step S10134 Update the token field according to a browsing or search request, and carry the updated token field in a browsing or search response message;
  • Step S1014 returning a browsing/search request response message to the user
  • Step S1015 Receive a request for a specified ring tone information sent by the user, and press the active path information of the specified ring tone information request into the stack sensitive information.
  • Step S10156 Update the token field according to the request for querying the ringtone information, and carry the updated token field in the response message for querying the specified ringtone information;
  • Step S1016 Return a specified ring tone information request response message to the user.
  • the present embodiment introduces a token (Token) for a malicious ringtone subscription by a browser address access and a legal login user through multiple submissions.
  • the ring mechanism filters each request message that is initiated. The message request is valid only for one session period, and the others are considered invalid operations.
  • the anti-malware subscription scheme of the class token ring mechanism is based on the stack tracking technology and then adds classes.
  • Token ring mechanism for example, by adding timestamps, MD5 algorithm, etc., on the one hand, to ensure that the current user always uses the newly generated token field.
  • the current token is generated by the token association of the previous operation. Therefore, it can be largely prevented from being sent after repeated transmission or tampering after the normal message is captured by the hacker.
  • the present embodiment further introduces a token-like ring mechanism, and generates a token field according to the request of the user mandatory step, and carries it in the corresponding response message, and returns it to the user according to the request of the subsequent user.
  • the card field is updated in association, which can effectively prevent the normal message from being repeatedly sent or falsified after being captured by the hacker, and further improves the security of the Web system, and ensures the CRBT user, the operator and the CRBT content provider.
  • an embodiment of the present invention provides a server for preventing a malicious ringing tone of a CRBT Web system, including: a stack module 401, a receiving module 402, a calculating module 403, and a subscription request processing module 404, where:
  • the stack module 401 is configured to record the active path information of the user in the process of ordering the ring tone, and push the active path information into the stack sensitive information;
  • the receiving module 402 is configured to receive a subscription request input by the user, and push the subscription request into the stack sensitive information;
  • the calculating module 403 is configured to calculate a credibility indicator of the order request by using a predetermined association iteration algorithm based on the stack sensitive information;
  • the order request processing module 404 is configured to filter out the order request when the credibility indicator exceeds a predetermined threshold.
  • the order request processing module 404 is further configured to allow the order request to pass when the confidence indicator does not exceed a predetermined threshold.
  • the malicious ring tone ordering problem caused by the program simulation robot impersonating a legitimate user to log in is introduced, and a path tracking technology of the stack mechanism is introduced, and the user is dynamically tracked in the CRBT Web system.
  • the activity status of the system, and through the status of these state changes to determine whether the current ordering operation is initiated by the user, the abnormal state behavior is regarded as an invalid ordering operation.
  • the ordinary CRBT user can be simplified from the login to the subscription behavior as the following key operations: login, browse/search, display, initiate a subscription based on the ring tone number, complete the order and prompt.
  • the illegal users only include some key operations such as ordering.
  • illegal users who are maliciously ordered usually use URL address shortcuts, or intercept and tamper with legitimate messages, and repeat the transmission to complete the order quickly, compared with normal ordinary users.
  • Required steps such as login, search, browse, etc. will not be executed, and the operations that initiate malicious ordering are performed very frequently, and the request parameters also conform to certain rules (because the software implementation does not really simulate the actual randomness).
  • the request such as the ringtone that is finally ordered, is likely to belong to only one SP, etc.).
  • this embodiment records technical steps such as a user requesting channel, such as adding a page verification code component, requesting an IP address, a user number, a current request page address, and the like in the login request, and recording the necessary steps of the user at the Web site.
  • the activity path pushes this information into the stack information so that it pops up when the last order operation is performed.
  • the algorithm of the class association rule is used to calculate the credibility indicator of the order request.
  • the legality of the user's order request is judged by the credibility indicator of the order request.
  • the credibility indicator of the order request is obtained by the association iterative algorithm.
  • the calculation formula of the association iterative algorithm is:
  • mandatory step x-1) refers to the mandatory step x-1
  • the conditional probability of the mandatory step X is executed.
  • the corresponding active path information needs to be extracted from the stack sensitive information to obtain a mandatory step for the user to log in to the ordering process.
  • conditional probability refers to the conditional probability that time A occurs for any event A and event B under the condition that the known time B occurs.
  • the formula for calculating the conditional probability is: P ( A
  • B ) P ( AB ) / P ( B );
  • P ( AB ) is the number of sample points contained in A under the condition of B
  • P ( B ) is the number of sample points under the condition of B occurrence.
  • the normal user's credibility P(A) 0.9 occurs in the login case;
  • the credibility of the normal user in the search action when logging in is:
  • the current operation is that the credibility of the normal user depends on the credibility of all the required steps.
  • the final abnormality order can be filtered out by a preset threshold.
  • the credibility indicator does not exceed the predetermined threshold, indicating that the subscriber is a normal subscriber, the subscription request is allowed to pass.
  • another embodiment of the present invention provides a server for preventing a malicious ringing tone by using a CRBT Web system. Based on the foregoing embodiments, the method further includes:
  • the token module 405 is connected between the stack module 401 and the receiving module 402, and configured to generate a token field according to the login request when the server receives the login request sent by the user, and carry the token field in the login response message. To the user.
  • the token module 405 is further configured to: when the server receives the browsing sent by the user or When the request is searched, the token field is updated according to the browsing or search request, and the updated token field is carried in a browsing or search response message and returned to the user.
  • the token module 405 is further configured to: when the server receives the request for querying the specified ringtone information sent by the user, update the token field according to the request for querying the ringtone information, and update the token The field is returned to the user in a response message for querying the specified ring tone information.
  • the present embodiment introduces a token (Token) for a malicious ringtone subscription by a browser address access and a legal login user through multiple submissions.
  • the ring mechanism filters each request message that is initiated. The message request is valid only for one session period, and the others are considered invalid operations.
  • the anti-malware subscription scheme of the Token Ring mechanism is based on the stack tracking technology and adds a token-like ring mechanism. For example, by adding timestamps and MD5 algorithms, the current user is always up-to-date.
  • the token field on the other hand, the current token is generated by the token association of the previous operation, and therefore, it can be largely prevented against repeated transmission or tampering after the normal message is captured by the hacker.
  • the method and the server for preventing the malicious ringing of the ring tones by the CRBT Web system by introducing a stack mechanism, recording the active path information of the mandatory step of the user in the process of ordering the ring tone, and pushing the active path information into the stack sensitive Information; based on the stack sensitive information, using a predetermined association iteration algorithm to calculate a credibility indicator of the subscription request; when the credibility indicator exceeds a predetermined threshold, filtering out the subscription request, thereby security for the CRBT web system
  • a token-ring-like mechanism is further introduced, and the token generated by the request of the user-required step is carried in the corresponding response message and returned to the user, and is requested according to the request of the subsequent user.
  • the card field is associated with the update, which can largely prevent the normal message from being sent after repeated transmission or tampering after being captured by the hacker.
  • security control is implemented on the way that the program simulating robot impersonates a legitimate user to log in, accesses through a browser address, legally logs in a user, and multiple submissions, such as multiple submissions, and improves the security of the Web system.
  • Security to a certain extent, guarantees the common interests between CRBT users, operators and CRBT content providers.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Security & Cryptography (AREA)
  • Computer Hardware Design (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Information Transfer Between Computers (AREA)

Description

彩铃 Web系统防止恶意订购铃音的方法及服务器 技术领域
本发明涉及通信技术领域， 尤其涉及一种浏览器端 /服务器 ( Browser/Sever, B/S )架构的彩铃 Web系统防止恶意订购铃音的方法及月良 务器。 背景技术
随着智能网业务的不断发展， 彩铃业务应用得越来越广泛。 其中， 以 Web服务器为基础的彩铃音展示、 试听、 订购及赠送等一系列操作均通过 Internet网络进行， 网络的开放性使得在此基础上的网络交互的安全性和有 效性则显得相当重要。
目前 B/S架构的彩铃 Web服务， 由于其开放的环境， 不可避免会受到 某些负面因素的影响， 其中恶意订购铃音是安全防范的一个重要内容， 但 是由于 Web服务的特性所限， 在此环节上的控制能力相对较弱， 随着彩铃 业务的深入开展，现有的 Web系统无法满足彩铃 Web服务的质量及安全性 需求。 发明内容
本发明的主要目的在于提供一种彩铃 Web系统防止恶意订购铃音的方 法及服务器， 以提高彩铃 Web系统的安全性。
为了达到上述目的， 本发明提出一种彩铃 Web系统防止恶意订购铃音 的方法， 包括：
服务器记录用户在订购铃音过程中必选步骤的活动路径信息， 并将所 述活动路径信息压入堆栈敏感信息； 接收用户输入的订购请求， 并将所述订购请求压入所述堆栈敏感信息； 基于所述堆栈敏感信息， 釆用预定的关联迭代算法计算所述订购请求 的可信度指标；
当所述可信度指标超过预定的阃值时， 过滤掉所述订购请求。
优选地， 所述必选步骤包括： 登录、 搜索和 /或浏览。
优选地，登录过程的活动路径信息包括页面验证码组件、请求 IP地址、 用户号码和 /或当前请求页面地址信息。
优选地， 该方法还包括： 当所述可信度指标未超过预定的阃值时， 允 许所述订购请求通过。
优选地， 该方法还包括：
当接收到用户发送的登录请求时， 服务器根据所述登录请求生成令牌 字段， 并将所述令牌字段携带在登录响应消息中返回给用户。
优选地， 在将登录响应消息返回给用户的步骤之后， 该方法还包括： 当接收到用户发送的浏览或搜索请求时， 服务器根据所述浏览或搜索 请求更新所述令牌字段， 并将更新后的所述令牌字段携带在浏览或搜索响 应消息中返回给用户。
优选地， 在将浏览或搜索响应消息返回给用户的步骤之后， 该方法还 包括：
当接收到用户发送的查询指定铃音信息的请求时， 服务器根据所述查 询指定铃音信息的请求更新所述令牌字段， 并将更新后的所述令牌字段携 带在查询指定铃音信息的响应消息中返回给用户。
本发明还提出一种彩铃 Web系统防止恶意订购铃音的服务器， 包括： 堆栈模块， 用于记录用户在订购铃音过程中必选步骤的活动路径信息， 并将所述活动路径信息压入堆栈敏感信息；
接收模块， 用于接收用户输入的订购请求， 并将所述订购请求压入所 述堆栈敏感信息；
计算模块， 用于基于所述堆栈敏感信息， 釆用预定的关联迭代算法计 算所述订购请求的可信度指标；
订购请求处理模块， 用于在所述可信度指标超过预定的阀值时， 过滤 掉所述订购请求。
优选地， 所述订购请求处理模块还用于， 在所述可信度指标未超过预 定的阀值时， 允许所述订购请求通过。
优选地， 还包括：
令牌模块， 用于在服务器接收到用户发送的登录请求时， 根据所述登 录请求生成令牌字段， 并将所述令牌字段携带在登录响应消息中返回给用 户。
优选地， 所述令牌模块还用于， 在服务器接收到用户发送的浏览或搜 索请求时， 根据所述浏览或搜索请求更新所述令牌字段， 并将更新后的所 述令牌字段携带在浏览或搜索响应消息中返回给用户。
优选地， 所述令牌模块还用于， 在服务器接收到用户发送的查询指定 铃音信息的请求时， 根据所述查询指定铃音信息的请求更新所述令牌字段， 并将更新后的所述令牌字段携带在查询指定铃音信息的响应消息中返回给 用户。
本发明提出的一种彩铃 Web系统防止恶意订购铃音的方法及服务器， 通过引入堆栈机制， 记录用户订购铃音过程中在 Web站点的必选步骤的活 动路径信息， 并将活动路径信息压入堆栈敏感信息； 基于该堆栈敏感信息， 釆用预定的关联迭代算法计算订购请求的可信度指标； 当可信度指标超过 预定的阀值时， 过滤掉该订购请求， 从而对彩铃 Web系统的安全性进行有 效控制； 同时， 在此基础上， 进一步引入类令牌环机制， 根据用户必选步 骤的请求生成令牌字段携带在相应的响应消息中返回给用户， 并根据后续 用户的请求对该令牌字段进行关联更新， 可以很大程度上有效防止对正常 消息经黑客手段捕获之后重复发送或篡改后发送。 通过本发明， 对程序模 拟机器人冒充合法用户登录、 通过浏览器地址访问、 合法登录用户通过多 次提交等多种恶意订购铃音的途径都进行了安全性控制， 提高了 Web系统 的安全性， 也一定程度上保证了彩铃用户、 运营商和彩铃内容提供商之间 的共同利益。 附图说明
图 1是本发明彩铃 Web系统防止恶意订购铃音的方法一实施例流程示 意图；
图 2是本发明彩铃 Web系统防止恶意订购铃音的方法一实施例中服务 器记录用户在订购铃音过程中必选步骤的活动路径信息， 并将活动路径信 息压入堆栈敏感信息的流程示意图；
图 3是本发明彩铃 Web系统防止恶意订购铃音的方法另一实施例中服 务器记录用户在订购铃音过程中必选步骤的活动路径信息， 并将活动路径 信息压入堆栈敏感信息的流程示意图；
图 4是本发明彩铃 Web系统防止恶意订购铃音的服务器一实施例结构 示意图；
图 5是本发明彩铃 Web系统防止恶意订购铃音的服务器另一实施例结 构示意图。 具体实施方式
本发明实施例解决方案主要思路是： 由于铃音恶意订购行为大多呈现 以下主要特征： 行为上的反常性， 比如普通用户很少或者几乎不可能去做 的行为； 行为上的规律性， 比如相同或呈某一序列的操作时间周期； 数据 上的规律性， 比如所有订购铃音编号的呈现自然序列关系等等， 因此， 本 实施例针对上述情形， 对于程序模拟机器人冒充合法用户进行登录而导致 的恶意铃音订购问题， 引入堆栈机制的路径跟踪技术； 对于浏览器地址访 问和合法登录用户通过多次提交而进行的恶意铃音订购， 引入类令牌环机 制的请求过滤器， 以对彩铃 Web系统的安全性进行有效控制。
以下实施例中的服务器为 Web服务器。
如图 1所示， 本发明一实施例提出一种彩铃 Web系统防止恶意订购铃 音的方法， 包括：
步骤 S101 , 服务器记录用户在订购铃音过程中必选步骤的活动路径信 息， 并将活动路径信息压入堆栈敏感信息。
本实施例针对程序模拟机器人冒充合法用户进行登录而导致的恶意铃 音订购问题， 引入堆栈机制的路径跟踪技术， 动态跟踪用户在彩铃 Web系 统上的活动状态， 并通过这些状态变化的情况来判断当前订购操作是否由 用户发起， 对异常的状态行为视为无效订购操作。
具体分析如下： 普通彩铃用户从登录到订购行为可简化描述为以下一 些关键操作： 登录、 浏览 /搜索、 展示、 根据铃音编号发起订购、 完成订购 并提示。 而非法用户则只包含订购等部分关键操作， 相比较可知， 恶意订 购的非法用户一般通过统一资源定位符（ URL, Universal Resource Locator ) 地址捷径， 或者截获并篡改合法消息， 重复发送而快速完成订购， 相对于 正常普通用户而言， 一些必选步骤如登录、 搜索、 浏览等操作均不会被执 行， 同时发起恶意订购的这些操作执行非常频繁， 而请求参数之间也符合 一定的规律， 因为软件的实现并不能真正模拟实际的随机请求， 比如最终 订购的铃音很可能仅归属一个 SP ( Service Provider, 服务提供商）等。
因此， 本实施例通过诸如用户请求渠道的技术限制， 比如在登录请求 中加入页面验证码组件、 请求 IP地址、 用户号码、 当前请求页面地址等敏 感信息， 记录用户在 Web站点的必经步骤的活动路径并将这些信息均压入 堆栈信息， 以便在最后订购操作时弹出， 釆用类关联规则的算法对订购请 求的可信度指标进行计算。
步骤 S102 , 接收用户输入的订购请求， 并将订购请求压入堆栈敏感信 息。
步骤 S103 , 基于堆栈敏感信息， 釆用预定的关联迭代算法计算订购请 求的可信度指标。
步骤 S104, 当可信度指标超过预定的阀值时， 过滤掉订购请求。
上述步骤 S103及步骤 S104中， 通过订购请求的可信度指标对用户的 订购请求的合法性进行判断。 其中， 订购请求的可信度指标通过关联迭代 算法计算获得。 关联迭代算法的计算公式为：
当前必选步骤 X是正常用户的可信度= P (必选步骤 x|必选步骤 X-1 ) P (必选步骤 x|必选步骤 x-1 )指的是必选步骤 x-1执行之后， 执行必 选步骤 X的条件概率。
在计算当前必选步骤 X是正常用户的可信度时， 需要从堆栈敏感信息 中提取相应的活动路径信息， 以获取用户从登录到订购过程中的必选步骤。
条件概率是指对任意事件 A和事件 B, 在已知时间 B发生的条件下时 间 A发生的条件概率。 条件概率的计算公式为：
P ( A|B ) = P ( AB ) / P ( B );
其中， P ( AB )是指在 B发生的条件下 A包含的样本点数； P ( B )是 指在 B发生的条件下样本点数。
以登录、 搜索和订购三个必选步骤为例， 设登录事件为 A, 设搜索事 件为 B,设订购事件为 C ,且同时设初始值正常用户登录事件发生的概率即 P ( A ) =0.9, 正常用户搜索事件发生的概率即 P ( B ) =0.8, 正常用户订购 事件发生的概率为 0.5。
对于必选步骤登录，在登录情况下发生的正常用户的可信度 P( A )=0.9; 对于必选步骤搜索， 在登录情况下搜索动作发生的正常用户的可信度 为： P ( B|A ) = P ( AB ) /P ( A ) = P ( B ) /P ( A ) =0.8/0.9 = 0.89;
同理， 对于必选步骤订购， 在搜索情况下发生的正常用户的可信度为： P ( C|B ) =P ( CB ) /P ( B ) = 0.5/0.89 = 0.56;
从上述计算公式可以看出， 当前操作是正常用户的可信度依赖于前面 所有必选步骤的可信度， 通过迭代计算， 随着路径长度的增加以及页面与 页面之间是否有跳转等关联关系， 越是正常用户， 该可信度指标越高， 反 之则越低， 最终可以通过预先设定的阔值过滤掉异常订购请求。
当可信度指标未超过预定的阀值时， 表明该订购用户为正常用户， 则 允许订购请求通过。
如图 2所示， 步骤 S101具体包括：
步骤 S1011 ,接收用户发送的登录请求， 记录登录请求中的页面验证码 组件、 请求 IP地址、 用户号码和 /或当前请求页面地址的活动路径信息， 并 将该登录请求的活动路径信息压入堆栈敏感信息；
步骤 S1012 , 向用户返回登录请求响应消息；
步骤 S1013 , 接收用户发送的浏览 /搜索请求， 并将浏览 /搜索请求的活 动路径信息压入堆栈敏感信息；
步骤 S1014, 向用户返回浏览 /搜索请求响应消息；
步骤 S1015 ,接收用户发送的指定铃音信息请求， 并将指定铃音信息请 求的活动路径信息压入堆栈敏感信息；
步骤 S1016, 向用户返回指定铃音信息请求响应消息。
本实施例通过引入堆栈机制 , 记录用户订购铃音过程中在 Web站点的 必选步骤的活动路径信息， 并将活动路径信息压入堆栈敏感信息； 基于该 堆栈敏感信息， 釆用预定的关联迭代算法计算订购请求的可信度指标； 当 可信度指标超过预定的阀值时， 过滤掉该订购请求， 从而对彩铃 Web系统 的安全性进行有效控制， 在提高系统安全性的同时， 也保证了彩铃用户、 运营商和彩铃内容提供商的共同利益。
如图 3所示， 本发明另一实施例提出一种彩铃 Web系统防止恶意订购 铃音的方法， 在上述实施例的基础上， 步骤 S101具体包括：
步骤 S1011 ,接收用户发送的登录请求， 记录登录请求中的页面验证码 组件、 请求 IP地址、 用户号码和 /或当前请求页面地址的活动路径信息， 并 将该登录请求的活动路径信息压入堆栈敏感信息；
步骤 S10112, 根据登录请求生成令牌字段， 并将令牌字段携带在登录 响应消息中；
步骤 S1012 , 向用户返回登录请求响应消息；
步骤 S1013 , 接收用户发送的浏览 /搜索请求， 并将浏览 /搜索请求的活 动路径信息压入堆栈敏感信息；
步骤 S10134, 根据浏览或搜索请求更新所述令牌字段， 并将更新后的 令牌字段携带在浏览或搜索响应消息中；
步骤 S1014, 向用户返回浏览 /搜索请求响应消息；
步骤 S1015 ,接收用户发送的指定铃音信息请求， 并将指定铃音信息请 求的活动路径信息压入堆栈敏感信息；
步骤 S10156, 根据查询指定铃音信息的请求更新令牌字段， 并将更新 后的令牌字段携带在查询指定铃音信息的响应消息中；
步骤 S1016, 向用户返回指定铃音信息请求响应消息。
本实施例与上述实施例的区别在于， 本实施例在上述实施例的基础上， 针对浏览器地址访问和合法登录用户通过多次提交而进行的恶意铃音订 购， 引入类令牌（Token )环机制， 对每次发起的请求消息进行过滤， 消息 请求仅在一个会话周期内有效， 其它均视为无效操作。
类令牌环机制的防恶意订购方案是在堆栈跟踪技术的基础上再加入类 令牌环机制， 比如通过加入时间戳、 MD5算法等技术手段， 一方面保证当 前用户使用的总是最新生成的令牌字段， 另一方面， 当前令牌是由上一操 作的令牌关联生成， 因此， 可以很大程度上有效防止对正常消息经黑客手 段捕获之后重复发送或篡改后发送。
本实施例在上述实施例的基础上， 进一步引入类令牌环机制， 根据用 户必选步骤的请求生成令牌字段携带在相应的响应消息中返回给用户， 并 根据后续用户的请求对该令牌字段进行关联更新， 可以很大程度上有效防 止对正常消息经黑客手段捕获之后重复发送或篡改后发送， 更进一步提高 了 Web系统的安全性， 保证了彩铃用户、 运营商和彩铃内容提供商之间的 共同利益。
如图 4所示， 本发明一实施例提出一种彩铃 Web系统防止恶意订购铃 音的服务器， 包括： 堆栈模块 401、 接收模块 402、 计算模块 403以及订购 请求处理模块 404, 其中：
堆栈模块 401 ,用于记录用户在订购铃音过程中必选步骤的活动路径信 息， 并将活动路径信息压入堆栈敏感信息；
接收模块 402, 用于接收用户输入的订购请求， 并将订购请求压入堆栈 敏感信息；
计算模块 403 , 用于基于堆栈敏感信息， 釆用预定的关联迭代算法计算 订购请求的可信度指标；
订购请求处理模块 404, 用于在可信度指标超过预定的阀值时， 过滤掉 订购请求。
订购请求处理模块 404还用于， 在可信度指标未超过预定的阀值时， 允许订购请求通过。
本实施例针对程序模拟机器人冒充合法用户进行登录而导致的恶意铃 音订购问题， 引入堆栈机制的路径跟踪技术， 动态跟踪用户在彩铃 Web系 统上的活动状态， 并通过这些状态变化的情况来判断当前订购操作是否由 用户发起， 对异常的状态行为视为无效订购操作。
具体分析如下： 普通彩铃用户从登录到订购行为可简化描述为以下一 些关键操作： 登录、 浏览 /搜索、 展示、 根据铃音编号发起订购、 完成订购 并提示。 而非法用户则只包含订购等部分关键操作， 相比较可知， 恶意订 购的非法用户一般通过 URL地址捷径， 或者截获并篡改合法消息， 重复发 送而快速完成订购， 相对于正常普通用户而言， 一些必选步骤如登录、 搜 索、 浏览等操作均不会被执行， 同时发起恶意订购的这些操作执行非常频 繁， 而请求参数之间也符合一定的规律（因为软件的实现并不能真正模拟 实际的随机请求， 比如最终订购的铃音很可能仅归属一个 SP等）。
因此， 本实施例通过诸如用户请求渠道的技术限制， 比如在登录请求 中加入页面验证码组件、 请求 IP地址、 用户号码、 当前请求页面地址等敏 感信息， 记录用户在 Web站点的必经步骤的活动路径并将这些信息均压入 堆栈信息， 以便在最后订购操作时弹出， 釆用类关联规则的算法对订购请 求的可信度指标进行计算。
通过订购请求的可信度指标对用户的订购请求的合法性进行判断。 其 中， 订购请求的可信度指标通过关联迭代算法计算获得。 关联迭代算法的 计算公式为：
当前必选步骤 X是正常用户的可信度 = P (必选步骤 x|必选步骤 X-1 ) P (必选步骤 x|必选步骤 x-1 )指的是必选步骤 x-1执行之后， 执行必 选步骤 X的条件概率。
在计算当前必选步骤 X是正常用户的可信度时， 需要从堆栈敏感信息 中提取相应的活动路径信息， 以获取用户从登录到订购过程中的必选步骤。
条件概率是指对任意事件 A和事件 B, 在已知时间 B发生的条件下时 间 A发生的条件概率。 条件概率的计算公式为： P ( A|B ) = P ( AB ) / P ( B );
其中， P ( AB )是指在 B发生的条件下 A包含的样本点数； P ( B )是 指在 B发生的条件下样本点数。
以登录、 搜索和订购三个必选步骤为例， 设登录事件为 A, 设搜索事 件为 B,设订购事件为 C ,且同时设初始值正常用户登录事件发生的概率即 P ( A ) =0.9, 正常用户搜索事件发生的概率即 P ( B ) =0.8, 正常用户订购 事件发生的概率为 0.5。
对于必选步骤登录，在登录情况下发生的正常用户的可信度 P( A )=0.9; 对于必选步骤搜索， 在登录情况下搜索动作发生的正常用户的可信度 为：
P ( B|A ) = P ( AB ) /P ( A ) =P ( B ) /P ( A ) =0.8/0.9 = 0.89;
同理， 对于必选步骤订购， 在搜索情况下发生的正常用户的可信度为：
P ( C|B ) =P ( CB ) /P ( B ) = 0.5/0.89 = 0.56;
即当前操作是正常用户的可信度依赖于前面所有必选步骤的可信度， 通过迭代计算， 随着路径长度的增加以及页面与页面之间是否有跳转等关 联关系， 越是正常用户， 该可信度指标越高， 反之则越低， 最终可以通过 预先设定的阔值过滤掉异常订购请求。
当可信度指标未超过预定的阀值时， 表明该订购用户为正常用户， 则 允许订购请求通过。
如图 5所示， 本发明另一实施例提出一种彩铃 Web系统防止恶意订购 铃音的服务器， 在上述实施例的基础上， 还包括：
令牌模块 405 , 连接在堆栈模块 401与接收模块 402之间， 用于在服务 器接收到用户发送的登录请求时， 根据登录请求生成令牌字段， 并将令牌 字段携带在登录响应消息中返回给用户。
进一步的， 令牌模块 405还用于， 在服务器接收到用户发送的浏览或 搜索请求时， 根据所述浏览或搜索请求更新所述令牌字段， 并将更新后的 所述令牌字段携带在浏览或搜索响应消息中返回给用户。
令牌模块 405还用于， 在服务器接收到用户发送的查询指定铃音信息 的请求时， 根据所述查询指定铃音信息的请求更新所述令牌字段， 并将更 新后的所述令牌字段携带在查询指定铃音信息的响应消息中返回给用户。
本实施例与上述实施例的区别在于， 本实施例在上述实施例的基础上， 针对浏览器地址访问和合法登录用户通过多次提交而进行的恶意铃音订 购， 引入类令牌（Token )环机制， 对每次发起的请求消息进行过滤， 消息 请求仅在一个会话周期内有效， 其它均视为无效操作。
类令牌环机制的防恶意订购方案是在堆栈跟踪技术的基础上再加入类 令牌环机制， 比如通过加入时间戳、 MD5算法等技术手段， 一方面保证当 前用户使用的总是最新生成的令牌字段， 另一方面， 当前令牌是由上一操 作的令牌关联生成， 因此， 可以很大程度上有效防止对正常消息经黑客手 段捕获之后重复发送或篡改后发送。
本发明实施例彩铃 Web系统防止恶意订购铃音的方法及服务器， 通过 引入堆栈机制， 记录用户订购铃音过程中在 Web站点的必选步骤的活动路 径信息， 并将活动路径信息压入堆栈敏感信息； 基于该堆栈敏感信息， 釆 用预定的关联迭代算法计算订购请求的可信度指标； 当可信度指标超过预 定的阀值时， 过滤掉该订购请求， 从而对彩铃 Web系统的安全性进行有效 控制； 同时， 在此基础上， 进一步引入类令牌环机制， 根据用户必选步骤 的请求生成令牌字段携带在相应的响应消息中返回给用户， 并根据后续用 户的请求对该令牌字段进行关联更新， 可以很大程度上有效防止对正常消 息经黑客手段捕获之后重复发送或篡改后发送。 通过本发明， 对程序模拟 机器人冒充合法用户登录、 通过浏览器地址访问、 合法登录用户通过多次 提交等多种恶意订购铃音的途径都进行了安全性控制， 提高了 Web系统的 安全性， 也一定程度上保证了彩铃用户、 运营商和彩铃内容提供商之间的 共同利益。
以上所述仅为本发明的优选实施例， 并非因此限制本发明的专利范围， 凡是利用本发明说明书及附图内容所作的等效结构或流程变换， 或直接或 间接运用在其它相关的技术领域， 均同理包括在本发明的专利保护范围内。

Claims

权利要求书
1、 一种彩铃网络 Web系统防止恶意订购铃音的方法， 其特征在于， 包 括：
服务器记录用户在订购铃音过程中必选步骤的活动路径信息， 并将所 述活动路径信息压入堆栈敏感信息；
接收用户输入的订购请求， 并将所述订购请求压入所述堆栈敏感信息； 基于所述堆栈敏感信息， 釆用预定的关联迭代算法计算所述订购请求 的可信度指标；
当所述可信度指标超过预定的阃值时， 过滤掉所述订购请求。
2、 根据权利要求 1所述的方法， 其特征在于， 所述必选步骤包括： 登 录、 搜索和 /或浏览。
3、 根据权利要求 2所述的方法， 其特征在于， 登录过程的活动路径信 息包括页面验证码组件、请求网络协议 IP地址、 用户号码和 /或当前请求页 面地址信息。
4、 根据权利要求 1所述的方法， 其特征在于， 该方法还包括： 当所述 可信度指标未超过预定的阃值时， 允许所述订购请求通过。
5、 根据权利要求 1-4中任一项所述的方法， 其特征在于， 该方法还包 括：
当接收到用户发送的登录请求时， 服务器根据所述登录请求生成令牌 字段， 并将所述令牌字段携带在登录响应消息中返回给用户。
6、 根据权利要求 5所述的方法， 其特征在于， 在将登录响应消息返回 给用户的步骤之后， 该方法还包括：
当接收到用户发送的浏览或搜索请求时， 服务器根据所述浏览或搜索 请求更新所述令牌字段， 并将更新后的所述令牌字段携带在浏览或搜索响 应消息中返回给用户。
7、 根据权利要求 6所述的方法， 其特征在于， 在将浏览或搜索响应消 息返回给用户的步骤之后， 该方法还包括：
当接收到用户发送的查询指定铃音信息的请求时， 服务器根据所述查 询指定铃音信息的请求更新所述令牌字段， 并将更新后的所述令牌字段携 带在查询指定铃音信息的响应消息中返回给用户。
8、一种彩铃 Web系统防止恶意订购铃音的服务器，其特征在于，包括： 堆栈模块， 用于记录用户在订购铃音过程中必选步骤的活动路径信息， 并将所述活动路径信息压入堆栈敏感信息；
接收模块， 用于接收用户输入的订购请求， 并将所述订购请求压入所 述堆栈敏感信息；
计算模块， 用于基于所述堆栈敏感信息， 釆用预定的关联迭代算法计 算所述订购请求的可信度指标；
订购请求处理模块， 用于在所述可信度指标超过预定的阀值时， 过滤 掉所述订购请求。
9、 根据权利要求 8所述的服务器， 其特征在于， 所述订购请求处理模 块还用于， 在所述可信度指标未超过预定的阃值时， 允许所述订购请求通 过。
10、 根据权利要求 8或 9所述的服务器， 其特征在于， 还包括： 令牌模块， 用于在服务器接收到用户发送的登录请求时， 根据所述登 录请求生成令牌字段， 并将所述令牌字段携带在登录响应消息中返回给用 户。
11、 根据权利要求 10所述的服务器， 其特征在于， 所述令牌模块还用 于， 在服务器接收到用户发送的浏览或搜索请求时， 根据所述浏览或搜索 请求更新所述令牌字段， 并将更新后的所述令牌字段携带在浏览或搜索响 应消息中返回给用户。
12、 根据权利要求 11所述的服务器， 其特征在于， 所述令牌模块还用 于， 在服务器接收到用户发送的查询指定铃音信息的请求时， 根据所述查 询指定铃音信息的请求更新所述令牌字段， 并将更新后的所述令牌字段携 带在查询指定铃音信息的响应消息中返回给用户。
PCT/CN2011/076205 2011-04-28 2011-06-23 彩铃web系统防止恶意订购铃音的方法及服务器 WO2012145962A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201110109049.7A CN102761576B (zh) 2011-04-28 2011-04-28 彩铃Web系统防止恶意订购铃音的方法及服务器
CN201110109049.7 2011-04-28

Publications (1)

Publication Number Publication Date
WO2012145962A1 true WO2012145962A1 (zh) 2012-11-01

Family

ID=47055897

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2011/076205 WO2012145962A1 (zh) 2011-04-28 2011-06-23 彩铃web系统防止恶意订购铃音的方法及服务器

Country Status (2)

Country Link
CN (1) CN102761576B (zh)
WO (1) WO2012145962A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103856920B (zh) * 2012-11-28 2017-09-12 中国移动通信集团河南有限公司 一种数据处理方法和装置
CN108289077B (zh) * 2017-01-09 2021-09-21 中兴通讯股份有限公司 一种对web服务器安全性进行模糊检测分析的方法及装置
CN109088999B (zh) * 2017-06-13 2021-04-06 中兴通讯股份有限公司 一种播放回铃音的方法、装置及计算机可读存储介质
CN110351259A (zh) * 2019-06-28 2019-10-18 深圳数位传媒科技有限公司 一种基于网络抓包获取app认证信息的方法及装置

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101217392A (zh) * 2007-12-27 2008-07-09 华为技术有限公司 行为采集分析方法及系统
CN101888619A (zh) * 2010-06-09 2010-11-17 中兴通讯股份有限公司 防止利用第三方交互语音响应平台恶意订购的方法及装置
CN101998372A (zh) * 2009-08-21 2011-03-30 中国移动通信集团广东有限公司 一种校验增值业务订购合法性的方法、设备及系统

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101257518B (zh) * 2008-03-27 2010-12-08 中国联合网络通信集团有限公司 防止wap平台中不经过计费网关的非法订购的方法及系统

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101217392A (zh) * 2007-12-27 2008-07-09 华为技术有限公司 行为采集分析方法及系统
CN101998372A (zh) * 2009-08-21 2011-03-30 中国移动通信集团广东有限公司 一种校验增值业务订购合法性的方法、设备及系统
CN101888619A (zh) * 2010-06-09 2010-11-17 中兴通讯股份有限公司 防止利用第三方交互语音响应平台恶意订购的方法及装置

Also Published As

Publication number Publication date
CN102761576A (zh) 2012-10-31
CN102761576B (zh) 2015-04-01

Similar Documents

Publication Publication Date Title
JP7144117B2 (ja) モデルトレーニングシステムおよび方法および記憶媒体
US9798896B2 (en) Managing unwanted tracking on a device
CN110855676B (zh) 网络攻击的处理方法、装置及存储介质
US8601586B1 (en) Method and system for detecting web application vulnerabilities
US20090100518A1 (en) System and method for detecting security defects in applications
WO2012113272A1 (zh) 一种提高终端上网安全性的方法、系统和装置
US20100205297A1 (en) Systems and methods for dynamic detection of anonymizing proxies
US8776240B1 (en) Pre-scan by historical URL access
CN109937564B (zh) 用于检测分布式计算系统中的欺诈性帐户使用的方法和设备
CN113259392B (zh) 一种网络安全攻防方法、装置及存储介质
WO2010105184A2 (en) A method and apparatus for phishing and leeching vulnerability detection
US9773068B2 (en) Method and apparatus for deriving and using trustful application metadata
US9336396B2 (en) Method and system for generating an enforceable security policy based on application sitemap
CA3094545A1 (en) Systems and methods of adaptively securing network communication channels
Stone-Gross et al. Peering through the iframe
JP2014534498A (ja) ＪａｖａＳｃｒｉｐｔを保護する装置、方法及びコンピューター可読性記憶媒体
US11178160B2 (en) Detecting and mitigating leaked cloud authorization keys
WO2013010394A1 (zh) 一种互联网病毒检测方法、装置和系统
Wu et al. Detect repackaged android application based on http traffic similarity
WO2012145962A1 (zh) 彩铃web系统防止恶意订购铃音的方法及服务器
CN116192371A (zh) 令牌发送、访问请求处理方法、装置、设备、介质及产品
CN107196811A (zh) 视频网站防盗链控制系统与方法
US11451564B1 (en) Systems and methods for threat disruption
CN114329611A (zh) 应用于联邦学习的权限管理方法、系统、装置及电子设备
CN112637171A (zh) 数据流量处理方法、装置、设备、系统和存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 11864451

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 11864451

Country of ref document: EP

Kind code of ref document: A1