Microsoft Technology Licensing, LLC

Anthony Paul Penta

Joshua T Goodman

Robert L Rounthwaite

Geoffrey John Hulten

Manav Mishra

Paul Stephen Rehfuss

Haitao Li

Nelly L Porter

Philip Joseph Hallin

Xiaohong Su

Kelvin Sjek Yiu

Elliott Jeb Haber

Muhammad Umar Janjua

Robert Alexander Sim

Aaron H Averbuch

Kenneth Grey Richards

Ryan Charles Colvin

Roderic C Deyo

Anooshiravan Saboori

Gopalakrishnan Seshadrinathan

David Maxwell Chickering

Eliot Carleton Gillum

Kelvin Shek Yiu

Christian Seifert

Bryston Mitsuo Nitta

Ameya S Bhatawdekar

Tomasz Kasperkiewicz

James Wesley Boerner

John Deurbrouck

David Aaron Ward Snelling

Kevin Patrick Kane

Jay T Buckingham

Vassil Dimitrov Bakalov

Zheng Dong

Zachariah Morgan

Roderict C Deyo

Anooshiravan Saboor

Michael Sitler

Tom Janes

Derk Adair Benisch

Jon Joy

Classifiers may be used to analyze a valid certificate received from an unverified entity in an attempt to establish a secure connection with the unverified entity. The classifiers may determine a probability that the certificate is being used improperly by an unauthorized third party. An action may be taken based on the probability, such as allowing the unverified entity to establish a secure connection, blocking the unverified entity from establishing a secure connection, etc. The classifiers may be trained by employing machine learning techniques on a collection of valid, authorized certificates. Additionally, or alternatively, received certificates may be sampled for further analysis based on the probability and/or predefined sampling percentages.

Probabilistic classifiers for certificates

Proxy networks enable a source to send traffic to one or more targets through a set of nodes operating as proxies. However, proxy networks are typically nonselective (often by design), and do not enable a source to specify properties of the nodes selected as proxies to send traffic to the target. Presented herein are proxy network techniques that enable sources to specify node properties in a target request, and that utilize a set of node managers for respective subsets of nodes. For a target request specifying selected node properties, the node managers may select as proxies nodes having the node properties specified in the target request. Additionally, the techniques presented herein promote the flexibility of the proxy network (e.g., adding groups of nodes, expanding the variety of nodes and node properties, and sending various forms of traffic using any protocol to various topics on behalf of many and various sources).

Scalable, selective traffic proxying

In many information security scenarios, a certificate issued by a certificate authority may be presented to a client in order to assert a trust level of a certificated item, such as a message or a web page. However, due to a decentralized structure and incomplete coordination among certificate authorities, the presence and exploitation of security vulnerabilities to issue untrustworthy certificates may be difficult to determine, particularly for an individual client. Presented herein are techniques for advising clients of the reputations of respective certificate authorities by evaluating the certificates issued by such certificate authorities, such as the number and types of domains certified by the certificate; the number and pattern of certificates issued for the domain; and the certification techniques used to issue the certificates. Such evaluation enables a determination of a certificate authority trust level that may be distributed to the clients in a certificate authority trust set.

Advising clients about certificate authority trust

In many information security scenarios, a certificate issued by a certificate authority on behalf of a domain is presented to a client in order to verify the identity of the domain. However, due to a decentralized structure and incomplete coordination among certificate authorities, the presence and exploitation of security vulnerabilities to issue untrustworthy certificates may be difficult for an individual client to determine. Presented herein are techniques for advising clients of the trustworthiness of respective certificate authorities by evaluating the certificates issued by such certificate authorities for suspicious indicators, such as hashcode collisions with other certificates and public key re-use. A trust level may be identified of respective certificate authorities according to the presence or absence of suspicious indicators in the certificates issued by the certificate authority, and a certificate authority trust set may be distributed to advise clients of the trustworthiness of certificates issued by the respective certificate authorities.

Certificate evaluation for certificate authority reputation advising

In many information security scenarios, a certificate issued by a certificating authority may be presented to a client in order to assert a trust level of a certificated item, such as a message or a web page. However, due to a decentralized structure and incomplete coordination among certificating authorities, the presence and exploitation of security vulnerabilities to issue untrustworthy certificates may be difficult to determine, particularly for an individual client. Presented herein are techniques for providing a certificating authority trust service that collects and evaluates certificates submitted to clients by certificating authorities, and advises the clients of a certificating authority trust level for respective certificating authorities (e.g., determined as a consensus of the evaluated certificates issued by the certificating authority). The clients may use a certificating authority trust set distributed by the certificating authority trust service to determine whether to trust a certificate issued from a particular certificating authority.

Certificating authority trust evaluation

One or more techniques and/or systems are provided for file acquisition for reputation evaluation. A reputation service may be configured to evaluate files and provide reputations of such files to clients (e.g., an indication as to whether a file is safe or malicious). If the reputation service receives a reputation request for a file that is unknown to the reputation service (e.g., a file not yet fully acquired by the reputation service), then the reputation service may identify a set of chunks into which the file can be partitioned. The reputation service may obtain chunks from various clients, such as a first chunk from a first client and a second chunk from a second client. Such chunks may be evaluated to assign a reputation to the file. In this way, the reputation service may retrieve portions of a file in a distributed manner for reputation evaluation.

Chunk-based file acquisition and file reputation evaluation

Malware detection is often based on monitoring a local application binary and/or process, such as detecting patterns of malicious code, unusual local resource utilization, or suspicious application behavior. However, the volume of available software, variety of malware, and sophistication of evasion techniques may reduce the effectiveness of detection based on monitoring local resources. Presented herein are techniques for identifying malware based on the reputations of remote resources (e.g., web content, files, databases, IP addresses, services, and users) accessed by an application. Remote resource accesses may be reported to a reputation service, which may identify reputations of remote resources, and application reputations of applications that utilize such remote resources. These application reputations may be used to adjust the application policies of the applications executed by devices and servers. These techniques thereby achieve rapid detection and mitigation of newly identified malware through application telemetry in a predominantly automated manner.

Identifying application reputation based on resource accesses

Described is a technology by which phishing-related data sources are processed into aggregated data and a given site evaluated the aggregated data using a predictive model to automatically determine whether the given site is likely to be a phishing site. The predictive model may be built using machine learning based on training data, e.g., including known phishing sites and/or known non-phishing sites. To determine whether an object corresponding to a site is likely a phishing-related object are described, various criteria are evaluated, including one or more features of the object when evaluated. The determination is output in some way, e.g., made available to a reputation service, used to block access to a site or warn a user before allowing access, and/or used to assist a hand grader in being more efficient in evaluating sites.

Finding phishing sites

In one embodiment, an intelligent detection systemmay determine if a network targetis an adversarial site based on comparing responses to different network sources. The intelligent detection systemmay select a test apparent network sourceand a control apparent network sourcefrom a network source pool. The intelligent detection systemmay receive the test response responding to a test request from the test apparent network sourceto a network target. The intelligent detection systemmay receive the control response responding to a control request from the control apparent network sourceto the network target. The intelligent detection systemmay execute a comparison of the test response to the control response.

Internet protocol address space management to mitigate and detect cloaking and other abuse

Phishing detection, prevention, and notification is described. In an embodiment, a messaging application facilitates communication via a messaging user interface, and receives a communication, such as an email message, from a domain. A phishing detection module detects a phishing attack in the communication by determining that the domain is similar to a known phishing domain, or by detecting suspicious network properties of the domain. In another embodiment, a Web browsing application receives content, such as data for a Web page, from a network-based resource, such as a Web site or domain. The Web browsing application initiates a display of the content, and a phishing detection module detects a phishing attack in the content by determining that a domain of the network-based resource is similar to a known phishing domain, or that an address of the network-based resource from which the content is received has suspicious network properties.

Phishing detection, prevention, and notification

Identification of email forwarders is described. In an implementation, a method includes using heuristics to identify email forwarders for use in a reputation system for locating spammers. In another implementation, a method includes determining a likelihood that a particular Internet Protocol (IP) address corresponds to an email forwarder and processing email originating from the particular IP address based on the determined likelihood. In a further implementation, a method includes collecting heuristic data that describes characteristics of emails sent from one or more Internet Protocol (IP) addresses and constructing a model from the heuristic data for identifying whether at least one of the IP address is an email forwarder. In yet a further implementation, a method includes identifying that a particular Internet Protocol (IP) address likely corresponds to an email forwarder and processing email from the particular IP address based on an implied sender of the email.

Identification of email forwarders

Disclosed are systems and methods that facilitate spam detection and prevention at least in part by building or training filters using advanced IP address and/or URL features in connection with machine learning techniques. A variety of advanced IP address related features can be generated from performing a reverse IP lookup. Similarly, many different advanced URL based features can be created from analyzing at least a portion of any one URL detected in a message.

Advanced URL and IP features

Growing community of inventors

Bellevue, WA, United States of America

Anthony Paul Penta