Abstract: A processor can be caused to receive, from a user compute device of a user, job data and execute a trained machine learning model using the job data to generate an estimation. The trained machine learning model can be trained using a training set including job descriptions correlated to values representing job descriptions. The processor is further caused to select available workers based on profiles associated with the available workers and stored in a database, to generate feature vectors for the available workers. The processor is further caused to determine a ranking for each available worker based on a difference between the job data and the feature vectors. The difference includes a distance between the locations of the user and that available worker. The processor is further caused to automatically assign to the job data the available worker having a highest ranking indicating the available worker has a shortest distance.

Abstract: The presently disclosed subject matter there includes a computerized device that comprises a processing circuitry, and one or more sensors including at least one image sensor configured to continuously capture images of an environment; the processing circuitry is configured to process the captured images and: detect an object within the environment; track a trajectory of movement of the object; select a best fitting motion model to the trajectory and determine a state of the object based on the best fitting motion model; and determine at least one intersection point between the object and a target object.

Abstract: A method of controlling a call between first and second user terminals, the method comprising: during the call, detecting a failure of a connection between the first user terminal and a packet-switched network; and in response, causing the call to be conducted at least part way via a PSTN network, via a connection between the first user terminal and the PSTN network. The call may be initiated by the first user terminal (such that the first user terminal is the caller and the second is the callee). The call may be initially conducted, prior to the failure, via a packet-switched connection between the first user terminal and a packet-switched network. The method may be implemented by a client application run on the first user terminal. Alternatively, the method may be implemented by a server.

Abstract: Method and system for classification in imbalanced datasets within a supervised classification framework. Bootstrap methodology is modified according to sampling weights and adaptive target set size principle, to induce weak classifiers from the bootstrap samples in an iterative procedure that results in a set of weak classifiers. A weighted combination scheme is used to adaptively combine the weak classifiers to a strong classifier that achieves good performance for all classes (reflected as high values for metrics such as G-mean and F-score) as well as good overall accuracy.

Abstract: A method of controlling a call between first and second user terminals, the method comprising: during the call, detecting a failure of a connection between the first user terminal and a packet-switched network; and in response, causing the call to be conducted at least part way via a PSTN network, via a connection between the first user terminal and the PSTN network. The call may be initiated by the first user terminal (such that the first user terminal is the caller and the second is the callee). The call may be initially conducted, prior to the failure, via a packet-switched connection between the first user terminal and a packet-switched network. The method may be implemented by a client application run on the first user terminal. Alternatively, the method may be implemented by a server.

Abstract: Disclosed is a communication system for effecting communication events between a plurality of endpoints connected via a communication network. The communication system comprises a plurality of processing units other than said endpoints, each having access to computer storage holding executable code modules for managing a communication event. The code modules are configured to implement a media modality controller configured to manage media modality of an established communication event and a call controller configured to establish the communication event. An instance of the call controller is assigned to progress the establishment of the communication event responsive to an instruction received via the network and is configured to initiate an instruction to at least one of: the media modality controller; and at least one of the endpoints.

Abstract: Disclosed herein is an instance of a media modality controller of a communication system which is assigned to convey media modality control signals of a communication event to respective media modality agents of endpoints of the communication event without accessing respective call agents of the endpoints. The media modality controller instance is so assigned independently of a call controller of the communication system and responsive to an instruction received via the network. The media modality controller instance is released from said assignment responsive to the media modality controller instance returning a response to the received instruction while the call controller continues to operate in communication with the call agents of the endpoints.

Abstract: Method and system for classification in imbalanced datasets within a supervised classification framework. Bootstrap methodology is modified according to k-Nearest Neighbor sampling weights and adaptive target set size principle, to induce weak classifiers from the bootstrap samples in an iterative procedure that results in a set of weak classifiers. A weighted combination scheme is used to adaptively combine the weak classifiers to a strong classifier that achieves good performance for all classes (reflected as high values for metrics such as G-mean and F-score) as well as good overall accuracy.

Abstract: Disclosed herein is an instance of a media modality controller of a communication system which is assigned to convey media modality control signals of a communication event to respective media modality agents of endpoints of the communication event without accessing respective call agents of the endpoints. The media modality controller instance is so assigned independently of a call controller of the communication system and responsive to an instruction received via the network. The media modality controller instance is released from said assignment responsive to the media modality controller instance returning a response to the received instruction while the call controller continues to operate in communication with the call agents of the endpoints.

Abstract: Disclosed is a communication system for effecting communication events between a plurality of endpoints connected via a communication network. The communication system comprises a plurality of processing units other than said endpoints, each having access to computer storage holding executable code modules for managing a communication event. The code modules are configured to implement a media modality controller configured to manage media modality of an established communication event and a call controller configured to establish the communication event. An instance of the call controller is assigned to progress the establishment of the communication event responsive to an instruction received via the network and is configured to initiate an instruction to at least one of: the media modality controller; and at least one of the endpoints.

Abstract: Disclosed is a first call controller instance of a communication system configured to access a first failure-tolerant region of computer storage to access a call state, the first call controller instance being assigned to so access the call state responsive to a first instruction received via a network. At least part of the call state is replicated in a second failure-tolerant region of the computer storage so that a second call controller instance of the communication system can access the at least part of the call state, the second call controller instance being assigned to so access the at least part of the call state responsive to a second instruction received via the network.