Abstract: Systems, methods, and computer-readable media for call classification and for training a model for call classification, an example method comprising: receiving DTMF information from a plurality of calls; determining, for each of the calls, a feature vector including statistics based on DTMF information such as DTMF residual signal comprising channel noise and additive noise; training a model for classification; comparing a new call feature vector to the model; predicting a device type and geographic location based on the comparison of the new call feature vector to the model; classifying the call as spoofed or genuine; and authenticating a call or altering an IVR call flow.

Abstract: Systems and methods for call detail record (CDR) analysis to determine a risk score for a call and identify fraudulent activity and for fraud detection in Interactive Voice Response (IVR) systems. An example method may store information extracted from received calls. Queries of the stored information may be performed to select data using keys, wherein each key relates to one of the received calls, and wherein the queries are parallelized. The selected data may be transformed into feature vectors, wherein each feature vector relates to one of the received calls and includes a velocity feature and at least one of a behavior feature or a reputation feature. A risk score for the call may be generated during the call based on the feature vectors.

Abstract: Systems, methods, and computer-readable media for call classification and for training a model for call classification, an example method comprising: receiving DTMF information from a plurality of calls; determining, for each of the calls, a feature vector including statistics based on DTMF information such as DTMF residual signal comprising channel noise and additive noise; training a model for classification; comparing a new call feature vector to the model; predicting a device type and geographic location based on the comparison of the new call feature vector to the model; classifying the call as spoofed or genuine; and authenticating a call or altering an IVR call flow.

Abstract: A torque converter including a cover arranged to receive torque; an impeller including an impeller shell non-rotatably connected to the cover; a turbine in fluid communication with the impeller and including a turbine shell; a first clutch axially displaceable to engage the cover; a second clutch axially displaceable to engage the impeller shell; and, a vibration damper including: a first input component non-rotatably connected to a component of the first clutch; a second input component non-rotatably connected to a component of the second clutch; and, an output flange arranged to non-rotatably connect to an input shaft for a transmission. For a first lock-up mode, the first and second clutches are arranged to close to transmit torque from the cover to the first and second input components, respectively.

Abstract: A clamp for holding a looped strap as required for a passenger assist strap on a railing in a public transit vehicle, comprising a first piece and a second piece with interlocking sides to hold the strap loop together and also prevent the first and second pieces from twisting out of alignment.

Abstract: Systems and methods for call detail record (CDR) analysis to determine a risk score for a call and identify fraudulent activity and for fraud detection in Interactive Voice Response (IVR) systems. An example method may store information extracted from received calls. Queries of the stored information may be performed to select data using keys, wherein each key relates to one of the received calls, and wherein the queries are parallelized. The selected data may be transformed into feature vectors, wherein each feature vector relates to one of the received calls and includes a velocity feature and at least one of a behavior feature or a reputation feature. A risk score for the call may be generated during the call based on the feature vectors.

Abstract: Systems and methods for call detail record (CDR) analysis to determine a risk score for a call and identify fraudulent activity and for fraud detection in Interactive Voice Response (IVR) systems. An example method may store information extracted from received calls. Queries of the stored information may be performed to select data using keys, wherein each key relates to one of the received calls, and wherein the queries are parallelized. The selected data may be transformed into feature vectors, wherein each feature vector relates to one of the received calls and includes a velocity feature and at least one of a behavior feature or a reputation feature. A risk score for the call may be generated during the call based on the feature vectors.

Abstract: The present invention relates to stanchion padding and more specifically to stanchion padding for use in transit vehicles. The stanchion padding of the present invention comprises a tube having a hollow centre running through it from a first end to a second end and the tube also comprising an exterior wall and an interior wall joined by at least two channel walls forming at least two hollow channels. These hollow channels may be adapted to receive end caps and at least one end cap may be installed onto either the first end or the second end of the tube. The interior wall may additionally have small inner channels running the length of the tube. The tube may be made by thermoplastic extrusion and at least one end cap may be made by thermoplastic injection molding.

Abstract: Systems, methods, and computer-readable media for call classification and for training a model for call classification, an example method comprising: receiving DTMF information from a plurality of calls; determining, for each of the calls, a feature vector including statistics based on DTMF information such as DTMF residual signal comprising channel noise and additive noise; training a model for classification; comparing a new call feature vector to the model; predicting a device type and geographic location based on the comparison of the new call feature vector to the model; classifying the call as spoofed or genuine; and authenticating a call or altering an IVR call flow.

Abstract: Systems and methods for call detail record (CDR) analysis to determine a risk score for a call and identify fraudulent activity and for fraud detection in Interactive Voice Response (IVR) systems. An example method may store information extracted from received calls. Queries of the stored information may be performed to select data using keys, wherein each key relates to one of the received calls, and wherein the queries are parallelized. The selected data may be transformed into feature vectors, wherein each feature vector relates to one of the received calls and includes a velocity feature and at least one of a behavior feature or a reputation feature. A risk score for the call may be generated during the call based on the feature vectors.

Abstract: A spring end cap includes a central axis, first and second parallel planar ends, and first and second circumferential surfaces. The planar ends are orthogonal to the central axis. An outer diameter of the first planar end is larger than an outer diameter of the second planar end. The first and second circumferential surfaces are disposed about the axis and between the planar ends. The first circumferential surface is disposed between the first planar end and the second circumferential surface. The first circumferential surface has a diameter smaller than a diameter of the second circumferential surface.

Abstract: Systems and methods for call detail record (CDR) analysis to determine a risk score for a call and identify fraudulent activity and for fraud detection in Interactive Voice Response (IVR) systems. An example method may store information extracted from received calls. Queries of the stored information may be performed to select data using keys, wherein each key relates to one of the received calls, and wherein the queries are parallelized. The selected data may be transformed into feature vectors, wherein each feature vector relates to one of the received calls and includes a velocity feature and at least one of a behavior feature or a reputation feature. A risk score for the call may be generated during the call based on the feature vectors.

Abstract: Systems and methods for call detail record (CDR) analysis to determine a risk score for a call and identify fraudulent activity and for fraud detection in Interactive Voice Response (IVR) systems. An example method may store information extracted from received calls. Queries of the stored information may be performed to select data using keys, wherein each key relates to one of the received calls, and wherein the queries are parallelized. The selected data may be transformed into feature vectors, wherein each feature vector relates to one of the received calls and includes a velocity feature and at least one of a behavior feature or a reputation feature. A risk score for the call may be generated during the call based on the feature vectors.

Abstract: A piston assembly for a torque converter is provided. The piston assembly includes a seal assembly including an inner ring configured for fixing to an outer circumferential surface of a transmission input shaft. The seal assembly also includes an outer ring on an outer circumferential surface of the inner ring. The piston assembly also includes a lock-up clutch piston including a radially inner end engaging the seal assembly. A method of forming a piston assembly for a torque converter is also provided. The method includes providing an outer ring on an outer circumferential surface of an inner ring to form a seal assembly configured for fixing to an outer circumferential surface of a transmission input shaft and connecting the seal assembly to a radially inner end of a lock-up clutch piston.

Abstract: A piston assembly for a transmission includes a piston housing and a release spring. The piston housing has a bore including a first radial wall and a circumferential wall. The release spring has a first radially outwardly extending finger portion, a second radially outwardly extending finger portion, and a conical portion. The first finger portion is for pressing engagement with the first radial wall and the second finger portion is for pressing engagement with the circumferential wall. The conical portion is disposed radially inward of the first and second finger portions. In an example embodiment, the piston housing includes a second radial wall for pressing against a clutch pack for the transmission. In some example embodiments, the piston assembly includes a seal plate, fixed to the piston housing and arranged to retain at least one piston seal.

Abstract: A piston assembly for a torque converter is provided. The piston assembly includes a seal assembly including an inner ring configured for fixing to an outer circumferential surface of a transmission input shaft. The seal assembly also includes an outer ring on an outer circumferential surface of the inner ring. The piston assembly also includes a lock-up clutch piston including a radially inner end engaging the seal assembly. A method of forming a piston assembly for a torque converter is also provided. The method includes providing an outer ring on an outer circumferential surface of an inner ring to form a seal assembly configured for fixing to an outer circumferential surface of a transmission input shaft and connecting the seal assembly to a radially inner end of a lock-up clutch piston.

Abstract: Tracking costs relative to a budget includes receiving one or more cost values for applying against a predefined budget, and comparing the one or more cost values against preset threshold percentages for the predefined budget, each preset threshold percentage indicating a percentage of costs applied against the predefined budget at which to notify users. Based on the comparison, a determination is made whether the one or more cost values meet or exceed at least one of the preset threshold percentages. In a case where the one or more cost values meet or exceed at least one of the preset threshold percentages, a notification is generated for a first user who bills against the predefined budget and for a second user who supervises the predefined budget, the notification indicating which of the preset threshold percentages are met or exceeded by the one or more cost values. The generated notification is provided to the first user and to the second user.

Abstract: A damper for a torque converter includes a first clutch arranged for selective engagement with a cover for the torque converter, a first spring, a second spring, and a first plate drivingly engaged with the first spring and the second spring. The damper also includes a second plate and a third plate. The second plate is axially displaceable relative to the first plate, drivingly engaged with one of the first or the second spring, and arranged for fixing to a turbine for the torque converter. The third plate is axially displaceable relative to the first and second plates, fixed to the first clutch, and drivingly engaged with the one of the first or second spring.