Abstract: This disclosure relates generally to a method and system for muting of classified information from an audio using a fuzzy approach. The method comprises converting the received audio signal into text using a speech recognition engine to identify a plurality of classified words from the text to obtain a first set of parameters. Further, a plurality of subwords associated with each classified word are identified to obtain a second set of parameters associated with each subword of corresponding classified word. A relative score is computed for each subword associated with the classified word based on a plurality of similar pairs for the corresponding classified word. A fuzzy muting function is generated using the first set of parameters, the second set of parameters and the relative score associated with each subword. The plurality of subwords associated with each classified word is muted in accordance with the generated fuzzy muting function.

Abstract: A method and a robotic system for online localized fatigue-state detection of a subject in a co-working environment using a non-intrusive approach is disclosed. A force sensor, mounted on the robotic system is capable of capturing effective force applied by local muscles of the subject co-working with the robotic system, providing a non-intrusive sensing. The captured force is analyzed on-line by the robotic system 102 to detect current fatigue state of the subject and proactively predict the future state of the subject. Thus, enables alerting the subject before time avoiding any possible accident.

Abstract: This disclosure relates generally to method and system for performing negotiation task using reinforcement learning agents. Performing negotiation on a task is a complex decision making process and to arrive at consensus on contents of a negotiation task is often expensive and time consuming due to the negotiation terms and the negotiation parties involved. The proposed technique trains reinforcement learning agents such as negotiating agent and an opposition agent. These agents are capable of performing the negotiation task on a plurality of clauses to agree on common terms between the agents involved. The system provides modelling of a selector agent on a plurality of behavioral models of a negotiating agent and the opposition agent to negotiate against each other and provides a reward signal based on the performance. This selector agent emulate human behavior provides scalability on selecting an optimal contract proposal during the performance of the negotiation task.

Abstract: Conversational systems are required to be capable of handling more sophisticated interactions than providing factual answers only. Such interactions are handled by resolving abstract anaphoric references in conversational systems which includes antecedent fact references and posterior fact references. The present disclosure resolves abstract anaphoric references in conversational systems using hierarchically stacked neural networks. In the present disclosure, a deep hierarchical maxpool network based model is used to obtain a representation of each utterance received from users and a representation of one or more generated sequences of utterances. The obtained representations are further used to identify contextual dependencies with in the one or more generated sequences which helps in resolving abstract anaphoric references in conversational systems.

Abstract: Disclosed is a system for identifying one or more web elements present on a web-page. A web-page receiving module may receive a web-page comprising a plurality of web elements. A parameter recording module may record one or more parameters corresponding to a web element of the plurality of web elements. A simulation module may simulate interaction of one or more input devices with the web element in order to determine behavior of the web element. The simulation module may further record change in the one or more parameters based on the interaction simulated. The change in the one or more parameters may be recorded upon triggering of at least one event. A web-element identification module may identify the web element based on the behavior of the web element and the change recorded in the one or more parameters.

Abstract: This disclosure relates generally to data structure abstraction, and more particularly to method and system for data structure abstraction for model checking. In one embodiment, the method includes identifying data structure accesses in the source code. Loops are identified in the data structure accesses, and loop-types are identified in the loops. An abstracted code is generated based on the loop types for abstracting the data structure. Abstracting the data structure includes, for each loop, replacing the data structure accesses by one of a corresponding representative element and a non-deterministic value in the loop body of said loop based on elements accessed, and eliminating loop control statement of said loop operating on elements of data structure based on loop type of said loop, and adding a plurality of non-array assignments at a start and after the loop body of the source code. The abstracted code is provided for the model checking.

Abstract: This disclosure relates generally to phase separation, and more particularly, to a apparatus and a method for phase separation. In one example, the apparatus includes a spiral shaped body, split outlets and an adjustable splitter. The spiral shaped body includes an inlet portion to receive a mixture of phases associated with distinct effective masses, an outlet portion, and multiple helical turns stacked between the inlet and outlet portion. A portion of helical turns are twisted to form a twisted portion having opposite walls of a preceding helical turn turned relative to one another in opposite directions. The split outlets are configured at walls of the preceding helical turn to withdraw the phases based on an effective mass of said phases. The adjustable splitter is movably configured at least a portion of a cross section of the spiral shaped body to facilitate separate withdrawal of the one or more phases of the mixture.

Abstract: Estimating Remaining Useful Life (RUL) from multi-sensor time series data is difficult through manual inspection. Current machine learning and data analytics methods, for RUL estimation require large number of failed instances for training, which are rarely available in practice, and these methods cannot use information from currently operational censored instances since their failure time is unknown. Embodiments of the present disclosure provide systems and methods for estimating RUL using time series data by implementing an LSTM-RNN based ordinal regression technique, wherein during training RUL value of failed instance(s) is encoded into a vector which is given as a target to the model. Unlike a failed instance, the exact RUL for a censored instance is unknown. For using the censored instances, target vectors are generated and the objective function is modified for training wherein the trained LSTM-RNN based ordinal regression is applied on an input test time series for RUL estimation.

Abstract: This disclosure relates generally to image processing, and more particularly to method and system for image reconstruction using deep dictionary learning (DDL). The system collects the degraded image as test image and processes the test image to extract sparse features from the test image, at different levels, using dictionaries. The extracted sparse features and data from the dictionaries are used by the system to reconstruct the HR image corresponding to the test image.

Abstract: This disclosure relates to systems and methods for solving generic inverse problems by providing a coupled representation architecture using transform learning. Convention solutions are complex, require long training and testing times, reconstruction quality also may not be suitable for all applications. Furthermore, they preclude application to real-time scenarios due to the mentioned inherent lacunae. The methods provided herein require involve very low computational complexity with a need for only three matrix-vector products, and requires very short training and testing times, which makes it applicable for real-time applications. Unlike the conventional learning architectures using inductive approaches, the CASC of the present disclosure can learn directly from the source domain and the number of features in a source domain may not be necessarily equal to the number of features in a target domain.

Abstract: The disclosure herein describes a method and a system for generating hybrid learning techniques. The hybrid learning technique refers to learning techniques that are a combination a plurality of techniques that include of deep learning, machine learning and signal processing to enable a rich feature space representation and classifier construction. The generation of the hybrid learning techniques also considers influence/impact of domain constraints that include business requirements and computational constraints, while generating hybrid learning techniques. Further from the plurality hybrid learning techniques a single hybrid learning technique is chosen based on performance matrix based on optimization techniques.

Abstract: A method and system for clustering users using cognitive stress report for classifying stress levels is provided. Detection and monitoring of cognitive stress experienced by users while performing a task is very crucial. The method includes receiving, user evaluated cognitive stress reports and the physiological signals of the user during the performance of the task. A normalized cognitive report is generated from the user evaluated cognitive stress report by computing mode and range value. The normalized cognitive stress reports of the users are used to cluster the users into a primary cluster and a secondary cluster. Feature sets are extracted from the physiological signals of the said users associated with the primary cluster. Using the said feature sets a classifier model is trained to classify the cognitive stress levels of the users as stressful class or stressless class.

Abstract: Traditional systems and methods have implemented hand-crafted feature extraction from varying length time series that results in complexity and requires domain knowledge. Building classification models requires large labeled data and is computationally expensive. Embodiments of the present disclosure implement learning models for classification tasks in multi-dimensional time series by performing feature extraction from entity's parameters via unsupervised encoder and build a non-temporal linear classifier model. A fixed-dimensional feature vector is outputted using a pre-trained unsupervised encoder, which acts as off-the shelf feature extractor. Extracted features are concatenated to learn a non-temporal linear classification model and weight is assigned to each extracted feature during learning which helps to determine relevant parameters for each class. Mapping from parameters to target class is considered while constraining the linear model to use only subset of large number of features.

Abstract: Method and system for automatic chromosome classification is disclosed. The system, alternatively referred as a Residual Convolutional Recurrent Attention Neural Network (Res-CRANN), utilizes property of band sequence of chromosome bands for chromosome classification. The Res-CRANN is end-to-end trainable system, in which a sequence of feature vectors are extracted from the feature maps produced by convolutional layers of a Residual neural networks (ResNet), wherein the feature vectors correspond to visual features representing chromosome bands in an chromosome image. The sequence feature vectors are fed into Recurrent Neural Networks (RNN) augmented with an attention mechanism. The RNN learns the sequence of feature vectors and the attention module concentrates on a plurality of Regions-of-interest (ROIs) of the sequence of feature vectors, wherein the ROIs are specific to a class label of chromosomes.

Abstract: Anomaly detection from time series is one of the key components in automated monitoring of one or more entities. Domain-driven sensor selection for anomaly detection is restricted by knowledge of important sensors to capture only a certain set of anomalies from the entire set of possible anomalies. Hence, existing anomaly detection approaches are not very effective for multi-dimensional time series. Embodiments of the present disclosure depict sparse neural network for anomaly detection in multi-dimensional time series (MDTS) corresponding to a plurality of parameters of entities. A reduced-dimensional time series is obtained from the MDTS via an at least one feedforward layer by using a dimensionality reduction model. The dimensionality reduction model and recurrent neural network (RNN) encoder-decoder model are simultaneously learned to obtain a multi-layered sparse neural network. A plurality of error vectors corresponding to at least one time instance of the MDTS is computed to obtain an anomaly score.

Abstract: Systems and methods for scheduling non-preemptive tasks in a multi-robot environment is provided. Traditional systems and methods facilitating preemptive task(s) allocation in a multi-processor environment are not applicable in the multi-robot environment since tasks are preemptive. Additionally, critical parameters like deadline and performance loss are not considered. Embodiments of the present disclosure provide for scheduling of a set of non-preemptive tasks by partitioning, the set of non-preemptive tasks either as a set of schedulable tasks or as a set of non-schedulable tasks; sorting, by a scheduling technique, the set of non-preemptive tasks partitioned; determining, by the scheduling technique, a possibility of execution of each of the set of schedulable tasks; and scheduling the set of schedulable tasks and the set of non-schedulable tasks upon determining the possibility of execution of each of the set of schedulable tasks.

Abstract: A method and system for generating Customer Decision Tree (CDT) for an entity in accordance with an attribute value (AV) based demand transfer estimation for a product category using machine learning, is disclosed. The method includes aggregating very high volume of data associated with a plurality of AVs of a product category at a plurality of aggregation levels. Further, generating a data matrix, which represents data is a structured format for machine learning, at a predefined aggregation level for the product category and generating a prediction model with the data matrix to determine predicted AV sales for each AV at the predefined aggregation level. Further, optimizing the trained prediction model. Thereafter, generate the CDT utilizing the optimized prediction model, a Demand Transfer (DT) estimator, a scenario generator and a hierarchy generator. Machine learning based DT is more accurate, effectively generating more accurate CDT tree.

Abstract: This disclosure relates generally to robotic network, and more particularly to a method and system for hierarchical decomposition of tasks and task planning in a robotic network. While a centralized system is used for action planning in a robotic network, any communication network issues can adversely affect working of the robotic network. Further, hardcoding one or more specific tasks to a robot restricts use of the robots irrespective of capabilities of the robots. The robotic agent decomposes a goal assigned to the robot to multiple sub-goals, and for each sub-goal, identifies one or more tasks to be executed/performed by the robot. An action plan is generated based on all such tasks identified, and the robot executes the action plan, in response to the goal assigned to the robot.

Abstract: Cloud robotics infrastructures generally support heterogeneous services that are offered by heterogeneous resources whose reliability or availability also varies widely with varying lifetime. For such systems, defining a static redundancy configuration for all services is difficult and often biased. Also, it is not feasible to define a redundancy configuration separately for each unique service. Therefore, in the present disclosure a trade-off between the two is ensured by providing At-most M-Modular Flexible Redundancy Model wherein an exact degree of redundancy is defined and is given to each service in a heterogeneous service environment and monitoring each task and subtask status to ensure that each subtask gets accomplished thereby enabling the tuning of the tradeoff between redundancy and cost and determining efficiency of the system by estimating number of resources utilized to complete specific subtask and comparing the resources utilization with the exact degree of redundancy defined.

Abstract: Traditionally arousal classification has been broadly done in multiple classes but have been insufficient to provide information about how arousal level of user changes over time. Present disclosure propose a continuous and unsupervised approach of monitoring the arousal trend of individual from his/her heart rate by obtaining instantaneous HR for time windows from a resampled time series of RR intervals obtained from ECG signal. A measured average heart rate (a measured HR) is computed from instantaneous HR specific to user for each time window thereby estimating apriori state based on a last instance of an aposteriori state initialized and observation of a state space model of Kalman Filter is determined for computing error and normalizing thereof which gets compared with a threshold for continuous monitoring of arousal trend of the user. The aposterior state is further updated using Kalman gain computed based on measurement noise determined for state space model.