Symantec Corporation

Ntt Docomo, Inc.

Cisco Technology, Inc.

Zulfikar Ramzan

Craig B Gentry

Sourabh Satish

Bernhard Bruhn

Oliver Friedrichs

Alfred Huger

Brian Hernacki

Philip Mackenzie

Shaun P Cooley

Carey S Nachenberg

Pratyusa Kumar Manadhata

Walter Bogorad

Mark Kevin Kennedy

Shane Pereira

Vadim Antonov

Adam J O'Donnell

Vijay Anand Seshadri

David P Woodruff

Candid Wüest

Ravi Kumar Jain

Govind Salinas

Toshiro Kawahara

Adam Glick

Sanjay Sawhney

David Alexander Molnar

Kent E Griffin

Pieter Viljoen

Vijay Seshadri

Adam Wright

Denis Serenyi

Chandrasekhar Kalle

Elias Levy

Alejandro Hevia

Ameet Zaveri

James Hoagland

Virendra Kumar

Andrea Del Miglio

Vijay Seshardi

Carey Nachenberg

Adam L Glick

Techniques are provided for the detection of malicious software (malware) on a general purpose computing device. A challenge in detecting malicious software is that files are typically scanned for the presence of malicious intent only once (and subsequent rescanning is typically performed in a simplistic manner). Existing methods in the art do not address how to most effectively rescan collections of files in a way that tries to optimize performance and efficacy. These methods may also be useful if additional information is now available regarding a file that might be useful to an end-user or an administrator, even though the file's core disposition might not have changed. More specifically, we describe methods, components, and systems that perform data analytics to intelligently rescan file collections for the purpose of retroactively identifying malware and retroactively identifying clean files.

Method and apparatus for retroactively detecting malicious or otherwise undesirable software as well as clean software through intelligent rescanning

A system retroactively detects malicious software on an end user system without performing expensive cross-referencing directly on the endpoint device. A client provides a server with information about files that are on it together with what it knows about these files. The server tracks this information and cross-references it against new intelligence it gathers on clean or malicious files. If a discrepancy in found (i.e., a file that had been called malicious, but that is actually benign or vice versa), the server informs the client, which in turn takes an appropriate action based on this information.

Method and apparatus for retroactively detecting malicious or otherwise undesirable software

A computer-implemented method for selectively authenticating a request based on an authentication policy is described. A request is received from a client. A determination is made as to which authentication threshold is applied to the request based on an authentication policy. The request is authenticated if the authentication threshold is satisfied. The authentication threshold is modified if the request is not successfully authenticated.

Systems and methods for selectively authenticating queries based on an authentication policy

The present invention relates to the security of general purpose computing devices, such as laptop or desktop PCs, and more specifically to the detection of malicious software (malware) on a general purpose computing device. A challenge in detecting malicious software is that files are typically scanned for the presence of malicious intent only once (and subsequent rescanning is typically performed in a simplistic manner). Existing methods in the art do not address how to most effectively rescan collections of files in a way that tries to optimize performance and efficacy. Accordingly we present novel methods, components, and systems for intelligently rescanning file collections and thereby enabling retroactive detection of malicious software and also retroactive identification of clean software. These methods may also be useful if additional information is now available regarding a file that might be useful to an end-user or an administrator, even though the file's core disposition might not have changed. More specifically, we describe methods, components, and systems that perform data analytics to intelligently rescan file collections for the purpose of retroactively identifying malware and retroactively identifying clean files. The disclosed invention provides a significant improvement with regard to efficacy and performance compared to previous approaches.

Reputations of objects are determined by a reputation system using reports from clients identifying the objects. Confidence metrics for the clients are generated using information determined from the reports. Confidence metrics indicate the amounts of confidence in the veracity of the reports. Reputation scores of objects are calculated using the reports from the clients and the confidence metrics for the clients. Confidence metrics and reputation scores are stored in correlation with identifiers for the objects. An object's reputation score is provided to a client in response to a request.

Using confidence about user intent in a reputation system

A server () uses a password (&#x3c0;) to construct a multiplicative group (Z*) with a (hidden) smooth order subgroup (&#x3c;x'&#x3e;), where the group order (P) depends on the password. The client () uses its knowledge of the password to generate a root extraction problem instance (z) in the group and to generate data (y) allowing the server to construct a discrete logarithm problem instance (y&#x2032;) in the subgroup. The server uses its knowledge of the group order to solve the root extraction problem, and solves the discrete logarithm problem efficiently by leveraging the smoothness of the subgroup. A shared key (sk) can be computed as a function of the solutions to the discrete logarithm and root extraction problem instances. In some embodiments, in an oblivious transfer protocol, the server queries the client (at) for data whose position in a database () is defined by the password. The client provides () such data without knowing the data position associated with the server's query. The client obtains the data position independently from the password. The data positions and/or the respective data are used for authentication and shared secret key generation. Other embodiments are also provided.

Cryptographic authentication and/or establishment of shared cryptographic keys, including, but not limited to, password authenticated key exchange (PAKE)

Descriptions of files detected at endpoints are submitted to a security server. The descriptions describe the names of the files and unique identifiers of the files. The security server uses the unique identifiers to identify files having different names at different endpoints. For a given file having multiple names, the names are processed to account for name differences unlikely to have been caused by malware. The processed names for the file are analyzed to determine the amount of dissimilarity among the names. This analysis is used to generate a score indicating a confidence that the computer file contains malicious software, where a greater amount of dissimilarity among the names generally indicates a greater confidence that the computer file contains malicious software. The score is weighted based on file name frequency, the age of the file, and the prevalence of the file. The weighted score is used to determine whether the computer file contains malicious software.

Malware detection using file names

To prevent gaming of a reputation system, a security token is generated for a security module using metadata about the client observed during the registration of the security module. The registration server selects metadata for use in generating the security token. The generated security token is provided to identify the client in later transactions. A security server may conduct a transaction with the client and observe metadata about the client during the transaction. The security server also extracts metadata from the security token. The security server correlates the observed metadata during the transaction with the extracted metadata from the security token. Based on the result of the correlation, a security policy is applied. As a result, the metadata in the security token enables stateless verification of the client.

Using metadata in security tokens to prevent coordinated gaming in a reputation system

A system for retroactively detecting malicious software on an end user system without performing expensive cross-referencing directly on the endpoint device. A client provides a server with information about files that are on it together with what it knows about these files. The server tracks this information and cross-references it against new intelligence it gathers on clean or malicious files. If a discrepancy is found (i.e., a file that had been called malicious, but that is actually benign or vice versa), the server informs the client, which in turn takes an appropriate action based on this information.

Application usage is profiled based on application streaming. Code pages of multiple applications are streamed from a server to multiple client computers (endpoints) for execution. The streaming of the code pages is monitored, and usage data is collected such as which pages are streamed to which endpoints, under what circumstances and when. By referencing the streamed code pages and the underlying source code, the code pages are mapped (at least approximately) to corresponding application features. The collected usage data usage and the relevant mapping are analyzed, to create application usage profile data for streamed applications. The application usage profile data can include such information as how often, when, where and by whom application components are being executed, as well as which components cause errors, are most popular, confuse users, etc.

Profiling application usage from application streaming

A method and apparatus for identifying domain name abuse in web-based content is described. In one embodiment, the method for identifying domain name abuse in web-based content to secure a computer comprising processing a first domain name and modifying the first domain name using a at least one rule for replacing characters. The modified first domain name indicates an imitation of a second domain name by the first domain name.

Method and apparatus for identifying domain name abuse

A computer-implemented method for identifying fraudulent websites. The method may include identifying a fraudulent-website toolkit. The fraudulent-website toolkit may be programmed for use in creating a fraudulent website. The method may also include determining a characteristic of the fraudulent-website toolkit and using the characteristic of the fraudulent-website toolkit to identify a website created using the fraudulent-website toolkit. The website created using the fraudulent-website toolkit may be identified by searching for websites that comprise the characteristic of the fraudulent-website toolkit. The website created using the fraudulent-website toolkit may also be identified by determining that a web browser is attempting to access the website. Various other methods and computer-readable media are also disclosed.

Methods and systems for identifying fraudulent websites

A candidate suspicious website is identified. A plurality of lightweight features associated with the candidate suspicious website is identified. A filter score is determined based on the plurality of lightweight features, wherein the filter score indicates a likelihood that the candidate suspicious website is a malicious website. Whether the filter score exceeds a threshold is determined. Responsive at least in part to the filter score exceeding the threshold it is determined that the candidate suspicious website is a suspicious website. Whether the suspicious website is a malicious website is determined by identifying software downloaded to the computing system responsive to accessing the suspicious website and determining whether the software downloaded to the computing system is malware based on characteristics associated with the downloaded software.

Filter-based identification of malicious websites

A pseudorandom number is generated from a random seed number using a collision-resistant hash function. A iteration input is extracted from the seed number. A hardcore predicate is applied to the iteration input to generate a pseudorandom bit. A pairwise-independent function is identified using a pairwise-independent function identifier extracted from the seed number and applied to the iteration input to produce a randomized iteration input. The collision-resistant hash function is applied to the randomized iteration input to produce a hash result and pad the output. The padded hash result is assigned as the iteration input for the next iteration. The process repeats iteratively and pseudorandom bits are generated using the hardcore predicate in each iteration until a predetermined number of pseudorandom bits are generated. The pseudorandom number is constructed using the generated pseudorandom bits.

Growing community of inventors

San Mateo, CA, United States of America

Zulfikar Ramzan