Abstract: Example apparatus and methods perform matrix factorization (MF) on a collaborative filter based usage matrix to create a multi-dimensional latent space that embeds users, items, and features. A full distance matrix is extracted from the latent space. The full distance matrix may be extracted from the latent space by defining a distance metric between item pairs based on the multi-dimensional representation in the latent space. The full distance matrix may be populated with values computed for item pairs using the distance metric. A plurality of vectors associated with a multi-dimensional Euclidean space are produced from the full distance matrix. The plurality of vectors produce a navigable data set. The plurality of vectors may be produced in a manner that minimizes strain on the distances vectors. A representation of the navigable data set may be presented as, for example, a virtually traversable landscape that supports an interactive user experience.

Abstract: A recommendation system is implemented using modified matrix factorization on top of a content-based matrix to provide both user-to-item and item-to-item content-based recommendations while exposing the full depth of transitive relationships among recommendations. Content information such as features and characteristics may be represented in a usage matrix in which features are treated as users would be in traditional matrix factorization. Matrix factorization is applied to the “features-as-users” matrix to build a content-based model in which features and items are embedded in a low dimension latent space. User history is employed for system training by locating user vectors within the latent space. Recommendations that are near to the vector can be provided to the users along with explanations (e.g., a recommendation is given because of an item's proximity to a particular feature).

Abstract: Example apparatus and methods transform a non-metric latent space produced by a matrix factorization process to a higher dimension metric space by applying an order preserving transformation to the latent space. The transformation preserves the order of the results of an inner product operation defined for the latent space. The higher dimension metric space may be queried for the results to different requests. Example apparatus and methods may assign every user i a vector ui in a latent space, and may assign every item j a vector vj in the latent space. The dot product ui·vj represents the score between the user i and the item j. The score represents the strength of the relationship between the user i and the item j. Example apparatus and methods may then apply ranking methodologies (e.g., LSH, K-D trees) to problems including recommendation, targeting, matchmaking, or item to item.

Abstract: Aspects of the subject matter described herein relate to a simplified login for mobile devices. In aspects, on a first logon, a mobile device asks a user to enter credentials and a PIN. The credentials and PIN are sent to a server which validates user credentials. If the user credentials are valid, the server encrypts data that includes at least the user credentials and the PIN and sends the encrypted data to the mobile device. In subsequent logons, the user may logon using only the PIN. During login, the mobile device sends the PIN in conjunction with the encrypted data. The server can then decrypt the data and compare the received PIN with the decrypted PIN. If the PINs are equal, the server may grant access to a resource according to the credentials.

Abstract: Methods, systems, and computer-readable media are disclosed for detecting vulnerabilities based on aggregated primitives. A particular method includes receiving a plurality of data transmissions. At least one of the data transmissions includes a protocol anomaly that is not indicative of a security threat. The method includes identifying a plurality of primitives associated with the data transmissions. The primitives are aggregated, and an attack condition is identified based on the aggregated primitives. A security alert is generated based on the identified attack condition.

Abstract: A recommendation system for optimizing content recommendation lists is disclosed. The system dynamically tracks a list interaction history of a user, which details that user's interactions with a plurality of different lists presenting different recommended items to that user. The system automatically correlates one or more list preferences with that user based on the list interaction history, and builds a recommendation list with a plurality of candidate items having different recommendation confidences. The recommendation list is built such that each candidate item with a higher recommendation confidence is prioritized over each candidate item with a lower recommendation confidence according to the one or more list preferences correlated to that user.

Abstract: Aspects of the subject matter described herein relate to a simplified login for mobile devices. In aspects, on a first logon, a mobile device asks a user to enter credentials and a PIN. The credentials and PIN are sent to a server which validates user credentials. If the user credentials are valid, the server encrypts data that includes at least the user credentials and the PIN and sends the encrypted data to the mobile device. In subsequent logons, the user may logon using only the PIN. During login, the mobile device sends the PIN in conjunction with the encrypted data. The server can then decrypt the data and compare the received PIN with the decrypted PIN. If the PINs are equal, the server may grant access to a resource according to the credentials.

Abstract: Example apparatus and methods perform matrix factorization (MF) on a usage matrix to create a latent space that describes similarities between users and items and between items and items in the usage matrix. The usage matrix relates users to items according to a collaborative filtering approach. A cell in the usage matrix may store a value that describes whether a user has acquired an item and the strength with which the user likes an item that has been acquired. The latent item space may reflect true relationships between items represented in the usage matrix and those relationships may be proportional to the strength in the usage matrix. The strength of the relationship may be encoded using continuous data that measures, for example, the amount of time a video game has been played, the amount of time content has been viewed, or other continuous or cumulative engagement measurements.

Abstract: Example apparatus and methods perform matrix factorization (MF) on a collaborative filter based usage matrix to create a multi-dimensional latent space that embeds users, items, and features. A full distance matrix is extracted from the latent space. The full distance matrix may be extracted from the latent space by defining a distance metric between item pairs based on the multi-dimensional representation in the latent space. The full distance matrix may be populated with values computed for item pairs using the distance metric. A plurality of vectors associated with a multi-dimensional Euclidean space are produced from the full distance matrix. The plurality of vectors produce a navigable data set. The plurality of vectors may be produced in a manner that minimizes strain on the distances vectors. A representation of the navigable data set may be presented as, for example, a virtually traversable landscape that supports an interactive user experience.

Abstract: Example apparatus and methods perform matrix factorization (MF) on a usage matrix to create a latent space that describes similarities between items in the usage matrix. The usage matrix relates source items that a user already has to target items that a user might acquire. A cell in the usage matrix may store a value that describes the likelihood (e.g., probability) that an acquisition of item x will lead to an acquisition of item y. The value stored in cell (x,y) is not transitive with the value stored in cell (y,x). Values that are missing in the usage matrix may be computed using vectors in the latent space. Once the usage matrix is updated, a directed recommendation may be produced from data in the usage matrix. Initial values in the usage matrix may be produced from data associated with actual acquisitions.

Abstract: Example apparatus and methods perform matrix factorization (MF) on a usage matrix to create a latent space that describes similarities between users and items in the usage matrix. The usage matrix relates users to items according to a collaborative filtering approach. A cell in the usage matrix may store a value that describes whether a user has acquired an item and the strength with which the user likes an item that has been acquired. Example apparatus and methods account for negative indications analytically rather than through negative sampling. Example apparatus and methods analyze strengths in the usage matrix, analyze item popularity, analyze user popularity, compute contribution factors for items with respect to users and users with respect to items, and compute new user vectors and new item vectors that depend on the strengths, popularity, and contributions. A recommendation may consider new user vectors and new item vectors.

Abstract: Example apparatus and methods transform a non-metric latent space produced by a matrix factorization process to a higher dimension metric space by applying an order preserving transformation to the latent space. The transformation preserves the order of the results of an inner product operation defined for the latent space. The higher dimension metric space may be queried for the results to different requests. Example apparatus and methods may assign every user i a vector ui in a latent space, and may assign every item j a vector vj in the latent space. The dot product ui·vj represents the score between the user i and the item j. The score represents the strength of the relationship between the user i and the item j. Example apparatus and methods may then apply ranking methodologies (e.g., LSH, K-D trees) to problems including recommendation, targeting, matchmaking, or item to item.

Abstract: The method of delegating authentication, within a chain of entities, relies upon a recording of at least a portion of a TLS handshake between a gateway device and user, in which the user needs access to a desired server. The method then relies upon re-verification of cryptographic evidence in the recorded portion of the TLS handshake, which is forwarded either (1) to the server to which access is desired, in which case the server re-verifies the recorded portion to confirm authentication, or, (2) to a third party entity, in which case the third party entity confirms authentication and provides credentials to the gateway server which then uses the credentials to authenticate to the server as the user.

Abstract: Matrix factorization techniques may be employed to identify different tastes based on user history information for a user profile and to provide item recommendations for the various tastes. An item model may be generated that includes item vectors, each item vector representing an item from a catalog of items. An item vector from the item model may be identified for each of a number of items identified in information for a user profile. The item vectors may be grouped into different clusters, and a taste vector may be generated for each cluster based on item vectors in each cluster. Each taste vector may be used to select item recommendations that may be combined in a set of recommendations provided for presentation to one or more users associated with the user profile.

Abstract: A technique for efficiently factoring a matrix in a recommendation system. Usage data for a large set of users relative to a set of items is provided in a usage matrix R. To reduce computational requirements, the usage matrix is sampled to provide a reduced matrix R?. R? is factored into a user matrix U? and an item matrix V. User vectors in U? and V are initialized and then iteratively updated to arrive at an optimal solution. The reduced matrix can be factored using the computational resources of a single computing device, for instance. Subsequently, the full user matrix U is obtained by fixing V and analytically minimizing an error in UV=R+error. The computations of this analytic solution can be divided among a set of computing devices, such as by using a map and reduce technique. Each computing device solves the equation for different respective subset of users.

Abstract: A client-based ad agent dynamically determines whether an advertisement campaign should bid on an impression for an end user and/or sets the bid price of the advertisement campaign for the impression. When an opportunity for an impression on a web page is identified, the ad agent accesses user data associated with an end user. The ad agent analyzes the user data to identify the relevance and/or value of serving an impression to the end user to the advertisement campaign. Based on the analysis, the ad agent controls whether the advertisement campaign bids on the impression for the end user and/or sets the bid price of the advertisement campaign for the impression.

Abstract: Disclosed herein is a system and method for identifying features of items that are more relevant for making recommendations to consumers for content that they may be interested in. The system determines the similarity between items that are recommend and items in the user's history and compares that similarity measure to the similarity measure calculated for a random item on the same features. From this similarity measure the relative impactfullness of a particular feature on a recommendation can be determined.

Abstract: Example apparatus and methods provide a recommendation to a user about a product they may wish to consider purchasing. One method produces a single indication concerning a relationship between a user and an item with which the user has interacted. The single indication identifies whether the user likes the item and the degree to which the user likes the item. The single indication is independent of user signals processed to compute the single indication. The single indication is produced by a signal deriver that is loosely coupled to a model of users and items. The model may be a matrix upon which matrix factorization can be performed. Although matrix factorization is performed, it is performed on vectors whose elements are independent of the signals processed by the signal deriver. Since users may have different preferences at different times, the degree to which the user likes the item may be manipulated.

Abstract: A single passcode can be used for validation by a user of several entities in a system without compromising security. The source of the entity providing validation credentials, along with the passcode, is considered when determining validity. A one-time password system validates credentials if a validation credentials, such as a user's valid passcode and the source of the credentials, have not been used previously. In a one-time passcode system, a validation processor receives validation credentials from a client processor. If the client processor has not previously sent the validation credentials to the validation processor, and the credentials are valid, the validation processor will validate the credentials. Otherwise, the credentials are invalid. Other client processors can utilize the same passcode and their respective source identifiers, and as long as the other client processors have not previously utilized the credentials, the credentials are declared valid.

Abstract: In a virtualization environment, a host machine on which a guest machine is operable is monitored to determine that it is healthy by being compliant with applicable policies (such as being up to date with the current security patches, running an anti-virus program, certified to run a guest machine, etc.) and free from malicious software or “malware” that could potentially disrupt or compromise the security of the guest machine. If the host machine is found to be non-compliant, then the guest machine is prevented from either booting up on the host machine or connecting to a network to ensure that the entire virtualization environment is compliant and that the guest machine, including its data and applications, etc., is protected against attacks that may be launched against it via malicious code that runs on the unhealthy host machine, or is isolated from the network until the non-compliancy is remediated.