Abstract: An approach for continual learning in slowly-varying environments is provided. The approach receives one or more action requests from a decision agent. The approach deploys a first model to a decision engine. The approach initiates an observation period. The approach builds a second model, in which the second model comprises collected transaction data from the observation period. The approach initiates a test period. The approach determines a performance score for the first model and a performance score for the second model. The approach selects the model providing an optimized action.

Abstract: An apparatus is provided for determining at least one of a domain name and an organization name associated with a server. The apparatus can include a traffic processor configured to acquire one or more handshake messages associated with establishing or resuming a secure session with the server. The apparatus can also include a site detector configured to determine whether the one or more handshake messages include one or more site textual identifiers. If the one or more handshake messages does not include one or more site textual identifiers, the site detector is configured to acquire the at least one of a domain name and an organization name based on querying a historical identification database using at least one of a session identifier and an IP address associated with the server.

Abstract: An apparatus is provided for classifying content in one or more transactions. The apparatus includes a transaction boundary detector configured to detect boundaries of the transactions associated with a first server or a first destination. The apparatus further includes a multimedia session generator configured to generate a multimedia session object based on the detected boundaries of the one or more transactions. The apparatus further includes a multimedia classifier configured to classify the content associated with a first multimedia session as being communicated under at least one of: HTTP live streaming (HLS) protocols, dynamic adaptive streaming over HTTP (DASH) protocols, adaptive bit rate protocols, progressive download protocols, or non-multimedia protocols.

Abstract: An apparatus is provided for estimating one or more quality of experience (QoE) parameters associated with a specific terminal. The apparatus includes a traffic processor configured to acquire a plurality of transactions for providing multimedia content to a specific terminal. At least one of such transactions is a secured transaction. The apparatus further includes a QoE parameter estimator configured to detect a quality level variation event based on the transactions and the sizes of the transactions, and to estimate one or more QoE parameters based on the detection of the quality level variation event.

Abstract: A system and method is provided for determining a client device identity. In one implementation, a method is provided that can include receiving a request from a client device. The method can also include determining, based on the client device request, a subscriber identification. After determining the subscriber identification, the method can include acquiring, from a transactional history database, transactional device data associated with the subscriber identification. In some embodiments, the transactional device data can include one or more counts associated with one or more device types. The method can also include determining, based on the transactional device data, the client device identity. In some embodiments, the determination of the client device identity can include selecting a device type of the one or more device types with the maximum count and setting the client device identity to the selected device type.

Abstract: An apparatus is provided for detecting the presence of multimedia content in one or more transactions and for classifying the multimedia content in the one or more transactions. The apparatus can include a traffic processor configured to acquire one or more handshake messages associated with the transactions. The apparatus can also include a multimedia detector configured to determine a domain name requested by a specific terminal based on the one or more handshake messages, and to detect the presence of the multimedia content data in the transactions using the determined domain name. The detection of the presence of the multimedia content can be used for at least one of optimizing or reporting of the multimedia content before the multimedia content is provided to the specific terminal.

Abstract: A method including acquiring a current HTTP transaction; determining whether the current HTTP transaction relates to web browsing for a specific client; acquiring a previous transactions set of the specific client; evaluating whether the current HTTP transaction belongs with the previous transactions set; if the current HTTP transaction belongs with the previous transactions set, adding the current HTTP transaction to the previous transactions set; and if the current HTTP transaction does not belong with the previous transactions set, creating a boundary of a page unit that includes the HTTP transactions of the previous transactions set for computing a page unit time.

Abstract: A system and method is provided for controlling multimedia data transmission to a communication device via a network. The system comprises a de-muxer configured to obtain a first multimedia data for transmission and a flow controller configured to evaluate an estimated buffer time associated with the communication device. If the estimated buffer time satisfies a first threshold condition, the flow controller updates a current transmitting state to be a pacing state. And if the estimated buffer time satisfies a second threshold condition, the flow controller updates the current transmitting state to be a bursting state. The system also comprises a muxer configured to transmit the first multimedia data at a transmitting rate corresponding to the bursting state or the pacing state.

Abstract: A system and method is provided for providing a composite Quality of Experience (QoE) metrics associated with a multimedia transmitted via a network. The method includes obtaining multimedia data corresponding to the multimedia transmitted via the network and acquiring metrics associated with the multimedia data. The metrics includes at least one multimedia transmitting metrics and at least one multimedia quality metrics. The method also includes normalizing the metrics and determining, based on the one or more normalized metrics, the composite multimedia QoE metrics.

Abstract: A system including a first user agent configured to transmit request data corresponding to a requested web page, wherein the request data includes a request data header, and an optimization server configured to receive the request data, to determine whether the request header data matches a classification of a grouping that identifies cached content data, wherein the grouping is configured based on similarities between a first content data and a second content data; and to transmit the cached content data to the user based on the determination.

Abstract: A method including acquiring a current HTTP transaction; determining whether the current HTTP transaction relates to web browsing for a specific client; acquiring a previous transactions set of the specific client; evaluating whether the current HTTP transaction belongs with the previous transactions set; if the current HTTP transaction belongs with the previous transactions set, adding the current HTTP transaction to the previous transactions set; and if the current HTTP transaction does not belong with the previous transactions set, creating a boundary of a page unit that includes the HTTP transactions of the previous transactions set for computing a page unit time.

Abstract: A method including determining whether historical content data is available for a time period; acquiring historical content data associated with a web content source if the historical content data is available; determining one or more user experience metrics for a specific terminal based on the acquired historical content data; and acquiring an expected quality of experience score based at least in part on the determined one or more user experience metrics, wherein the expected quality of experience score can be used for providing one or more traffic management optimization rules for current content data to the specific terminal.

Abstract: A computer-implemented method including providing, at a first device, a packet scheduler layer between a network layer and a transport layer; receiving, at the packet scheduler layer, one or more transmission control protocol (TCP) packets from a sending layer on the first device, wherein the sending layer is one of the network layer or the transport layer; smoothing delivery of at least one of the one or more TCP packets by delaying the delivery; and sending the one or more TCP packets to a receiving layer, wherein the receiving layer is one of the network layer or the transport layer that is not the sending layer.

Abstract: A computer-implemented method including providing, at a first device, a packet scheduler layer between a network layer and a transport layer; receiving, at the packet scheduler layer, one or more transmission control protocol (TCP) packets from a sending layer on the first device, wherein the sending layer is one of the network layer or the transport layer; smoothing delivery of at least one of the one or more TCP packets by delaying the delivery; and sending the one or more TCP packets to a receiving layer, wherein the receiving layer is one of the network layer or the transport layer that is not the sending layer.

Abstract: A system including a first user agent configured to transmit request data corresponding to a requested web page, wherein the request data includes a request data header, and an optimization server configured to receive the request data, to determine whether the request header data matches a classification of a grouping that identifies cached content data, wherein the grouping is configured based on similarities between a first content data and a second content data; and to transmit the cached content data to the user based on the determination.

Abstract: A method of representing handwriting which includes receiving a pen input, for example at a digitizer (10), segmenting the input into strokes (13), grouping the strokes into stroke sub-structures (14) and quantizing the stroke sub-structures according to a predefined set (the “alphabet”) of stroke sub-structures. For handwriting recognition, distance measurements are computed between the input stroke sub-structures and members of a predefined set or “alphabet” (19) of stroke sub-structures.

Abstract: The present invention provides a user interface having a plurality of user entry regions (34 and 36) Each user entry region is preferably separated into a first user entry sub-region (40) and a second user entry sub-region (42) by a visual cue (38). A first character is received in the first user entry sub-region (40) and a second character is received in the second user entry sub-region (42) to provide a unitary collection of strokes. The unitary collection of strokes is then matched as a whole against at least one character template in order to identify a desired character of a user.

Abstract: A storage medium (72) having stored thereon a set of instructions, which when loaded into a microprocessor (74), causes the microprocessor (74) to extract strokes from a plurality of characters (76), derive a pre-defined number of stroke models based on the strokes extracted from the plurality of character (78) and represent the plurality of characters as sequences of stroke models (80).

Abstract: A method and apparatus for implementing a graphical user interface keyboard (10) and a text buffer (12) on an electronic device. A character that is active upon pointer-up is accepted as a text character, even though the character that is active upon pointer-up is different from a character that was active and inserted in the text buffer (12) upon pointer-down.

Abstract: A method and apparatus for recognition of handwritten input is disclosed where handwritten input composed of a sequence of (x, y, pen) points, is preprocessed into a sequence of strokes. A short list of candidate characters that are likely matches for the handwritten input is determined by finding a fast matching distance between the input sequence of strokes and a sequence of strokes representing each candidate character of a large character set where the sequence of strokes for each candidate character is derived from statistical analysis of empirical data. A the final sorted list of candidate characters which are likely matches for the handwritten input is determined by finding a detailed matching distance between the input sequence of strokes and the sequence of strokes for each candidate character of the short list. A final selectable list of candidate characters is presented to a user.