Abstract: This disclosure relates generally to a method and system for damage localization on surfaces made of composites and metals. State-of-the-art methods for ultrasonic guided wave-based damage localization provide a reasonable accuracy. However, accuracy of prediction based on minimum number of observations is not yet achieved. The disclosed method provides damage localization by capturing response to the ultrasonic tone burst transmitted by a plurality of active piezoelectric sensors. The disclosed method provides a modified RAPID algorithm that considers an attenuation of the ultrasonic guided waves and factors energy of transmitted and received signals while predicting damage location. The method provides iterative grid search reduction mechanism to predict damage on the surfaces made of composites and metals.

Abstract: The disclosure relates generally to methods and systems for real-time voltage stabilization of electrical distribution networks with non-linear power flows. Existing real-time voltage control techniques are neither performance nor stability guarantees. The present disclosure proposes an online robust control algorithm, which operates without knowing an exact information of the line-parameters and resolves the voltage stability problem. In the proposed method a load data, a distributed energy resources (DER) data, and a network data of an electrical distribution network is obtained, to obtain a voltage profile at each time-step of the electrical distribution network. Next, line-parameters of the electrical distribution network are predicted using an on-line convex optimization technique and a Gauss-Seidel technique.

Abstract: Developability of a drug candidate is decided based on the Pharmacokinetic (PK) and Pharmacodynamic (PD) parameters of the drug candidate under investigation. Present disclosure provides systems and methods that are implemented using universal PK parameters' bounds and optimization technique(s) to produce robust and optimized set of PK parameters. More specifically, the system and method for estimating optimized set of PK parameters by a) creating universal parameter bounds, b) performing logical operations on universal PK parameters' bounds to create multiple bound combinations c) computing a performance threshold for residual sum of squares (RSS) d) performing global optimization to estimate globally optimized set of PK parameters act as initial PK parameters and e) performing local optimization of initial PK parameters to estimate locally optimized set of PK parameters, the best PK parameters that can used for assessing the developability of drug candidates within Pharmaceutical industry.

Abstract: Arrival/Travel times for public transit exhibit variability on account of factors like seasonality, dwell times at bus stops, traffic signals, travel demand fluctuation, spatial and temporal correlations, etc. The developing world in particular is plagued by additional factors like lack of lane discipline, excess vehicles, diverse modes of transport and so on. This renders the bus arrival time prediction (BATP) to be a challenging problem especially in the developing world. Present disclosure provides system and method that implement recurrent neural networks (RNNs) for BATP (in real-time), wherein the system incorporates information pertaining to spatial and temporal correlations and seasonal correlations. More specifically, a Gated Recurrent Unit (GRU) based Encoder-Decoder (ED) model with one or more bi-directional layers at the decoder is implemented for BATP based on relevant additional synchronized inputs (from previous trips) at each step of the decoder.

Abstract: The disclosure relates generally to methods and systems for transforming qualitative survey into quantitative survey. Current approaches depend on manual analysis of these user responses which is so troublesome task. The present disclosure transforms the qualitative survey questionnaire into the quantitative survey questionnaire using a domain knowledge and a natural language processing. The method first receives responses to each question present in qualitative survey questionnaire, from multiple batches. Then valid responses out of all the responses are determined for each question, pertaining to each batch, using domain taxonomy and natural language knowledge graph. Further, semantic relation-based technique is employed to determine the questions that are transformable batch wise. Then, the response options are created for each transformable question.

Abstract: A method and system for task planning for visual room rearrangement under partial observability is disclosed. The system or the robotic agent utilizes a visual input to efficiently plan a sequence of actions for simultaneous object search and rearrangement in an untidy room, to achieve a desired tidy state. Unlike search networks in the art that follow ad hoc approach, the method discloses a search network utilizing commonsense knowledge from large language models to find unseen objects. A Deep RL network used for task planning is trained with proxy reward, along with unique graph-based state representation to produce a scalable and effective planner that interleaves object search and rearrangement to minimize the number of steps taken and overall traversal of the agent, and to resolve blocked goal and swap cases. Sample efficient cluster-biased sampling is utilized for simultaneous training of the proxy reward network along with the Deep RL network.

Abstract: HVAC control system's supervisory control is crucial for energy-efficient thermal comfort in buildings. The control logic is usually specified as ‘if-then-that-else’ rules that capture the domain expertise of HVAC operators, but they often have conflict that may lead to sub-optimal HVAC performance. Embodiments of the present disclosure provide a method and system for optimized Heating, ventilation, and air-conditioning (HVAC) control using domain knowledge combined with Deep Reinforcement Learning (DRL). The system disclosed utilizes Deep Reinforcement Learning (DRL) for conflict resolution in a HVAC control in combination with domain knowledge in form of control logic. The domain knowledge is predefined in an Expressive Decision Tables (EDT) engine via a formal requirement specifier consumable by the EDT engine to capture domain knowledge of a building for the HVAC control.

Abstract: The taxonomic resolution obtained with conventional sequencing methods like Sanger (longer read lengths) takes a huge amount of time. While, NGS technologies (shorter read lengths) involves a lot of cost in sequencing. In addition to that the accuracy and depth of taxonomic classification is also less. A method and system for improving accuracy of amplicon sequencing based taxonomic profiling of microbial communities has been provided. The proposed strategy relies on obtaining taxonomic abundance profiles of a microbial community from two paired-end sequencing experiments, each of which targets different pair-wise combinations of non-contiguous (or contiguous) V-regions. The two taxonomic profiles are then combined based on (pre-estimated) accuracies of the individual V-regions (targeted in the experiments) in resolving each of the taxonomic groups under consideration.

Abstract: Continuous monitoring of subject's cardiac system using biological signal(s) (BS) during day-to-day activities is essential for managing personal cardiac health/disorders, etc. Conventional systems/methods lack in improvising overall classification results and configured for specific device/signal say ECG or PPG and so on. Present disclosure provides systems and methods for classifying BS obtained from users, wherein BS are preprocessed to obtain filtered signals (FS). Corresponding feature extraction module is utilized for feature set extraction based on features in FS. The feature set is reduced and segmented into test and training data. Biological signal classification model(s) are generated using training data and a BCM is applied on test data to classify biological signals (BS) as one of Atrial Fibrillation (AF), a non-AF, a cardiac arrythmia disorder, or ischemia.

Abstract: This disclosure relates to field of cryptography and digital signatures. The constant theft of cryptocurrencies is due to compromise of the secret signing key inherently stored in a single location. Also, one of challenges in the existing ECDSA signature is single point failure, wherein the signing key (private key) is prone to theft. The disclosed technique overcomes the challenging in the existing techniques by party distributed signature that ensures the safety of the private key. The disclosed techniques slits the key in two parts and saves at two different locations/machines. Further based on a ECDSA based technique, the digital signature is obtained securely using several steps that includes generation of first two parts of digital signature at one party, generation of the second two parts of digital signature at second party, finally decrypting a complete digital signature at the first party.

Abstract: The present disclosure discloses a method and system for generating recommendations for cloud instances for high performance computing (HPC) applications. The present disclosure provides an intelligent Cloud instance Recommender framework comprising a suitability matcher, a performance analyzer, and a decision making enabler. The method of the present disclosure ensures that the HPC application is assessed for its suitability for the cloud since there is no need of recommending cloud services if the HPC application cannot be migrated to the cloud. This assessment is performed using a machine learning (ML) predictor engine which is trained upon some parameters of the HPC application. The ML predictor engine predicts execution time of the HPC application on cloud instances, and then a cost of execution is estimated by a mathematical model based on the predicted execution time. Also, a weightage to user's input is provided using a recommender engine to generate final recommendations.

Abstract: Existing techniques for automated generation of test data for testing web applications need detailed requirement documents. The present disclosure receives a plurality of textual documents to extract context. Rephrasing the extracted context by implementing a plurality of rules and passing extracted context along with a first set of prompts to Large Language Model (LLM). Generating program, validator and first set of constraints for extracted context and generating test data by running the generated program. Assigning ranking to test data and selecting the test data with highest ranking. Statically refining the generated program by calling a mathematical library function on the highest ranked test data to generate structural information and modifying language of the second set of prompts passed to the LLM. Dynamically refining the generated program by passing feedback generated by executing the highest ranked test data on a web application and refining the response obtained.

Abstract: Deep learning-based generative models have improved the exploration of chemical space in small molecule drug discovery. Although thousands of novel small molecules can be generated with such models, synthesizing them still remains a challenging task. In literature, several methods have been proposed to predict the synthetic route of a target molecule by working backwards to find the most suitable starting reactants (retrosynthesis). While retrosynthesis is shown to be successful, for novel molecules it is often difficult to find the synthesis path. System and method of the present disclosure generate molecules along with its synthesis route and also provide an insight into the interactions in the active site of target protein, using graph convolution networks (GCNs) and Monte Carlo tree search (MCTS). A target-specific bioactivity prediction model is used as the scoring function to navigate the MCTS search space efficiently.

Abstract: This disclosure relates generally to a method and system for generative Al based unified virtual assistant. Conventional virtual assistant for enterprise systems needs to be configured for a specific industry or stakeholder and does not provide support for all stakeholders in the enterprise. Also, conventional rule-based virtual assistant or machine learning based virtual assistant need a large database for proper functioning. The disclosed method and system provide a unified virtual assistant for all processes in the enterprise. The unified virtual assistant provides support for all stakeholders in the enterprise and can answer all kinds of queries related to any process of the enterprise according to a role of a user logged into the system. The unified virtual assistant interprets user's query and generates effective prompts depending on the user's query which can be specific to customer, employee, executive or support desk users of the enterprise.

Abstract: This disclosure relates generally to shelf life of produce and, more particularly, for predicting and enhancing shelf life of produce in storage facility. A significant quantity of produce such as fresh fruits and vegetable are lost before reaching the consumer, during its long-term storage in a warehouse or a storage facility. Many techniques have been employed to preserve-enhance the shelf life. However, the existing techniques do not explicitly consider factors such as air circulation, stacking of container, and respiration of the produce during shelf-life prediction. The disclosed techniques predict and enhance the shelf life of produce in storage facilities in several steps including determining a set of modelling parameters, determining a plurality of shelf-life parameters, predicting a shelf life of the produce based on generating a shelf-life prediction model, predicting a quality index and finally, enhancing the shelf-life of the produce based on an optimization technique.

Abstract: This disclosure relates generally to method and system for monitoring human parameters using hierarchical human activity sensing. The method is based on sensing as service (SEAS) model which processes continuous mobility data from multiple sensors on the client edge-device by optimizing the on-device processing pipelines. The method requests a subject to select a human parameter of the human body to be monitored using a master device and capture the plurality of signals by recognizing sensors corresponding to the health parameter. The master device transmits to the server the subject selected human parameter of the human body to be monitored and requesting the server to recommend a hierarchical classifier structure. Further, the human body is monitored based on the on-device hierarchical sensing pipeline by executing a plurality of algorithms. In addition, the system is suitable for remote monitoring and flexible edge cloud arbitration, optimizing costs, infrastructure, and energy.

Abstract: This disclosure relates generally to a method and system for re-encryption of an encrypted data. State-of-the-art methods provide the re-encryption scheme for a specific Fully Homomorphic Encryption (FHE) encrypted data. However, a generic scheme that converts any given FHE scheme to an HPRE scheme is not yet achieved. The disclosed method provides re-encryption of the encrypted data transferred between a first user and a second user by a re-encryption key. The re-encryption key is obtained by splitting a private key of the first user into a primary private key and a secondary private key. The primary private key generates a public re-key component using probabilistic encryption algorithm; and the secondary private key generates a private re-key component using probabilistic switch key generation algorithm. Both the private re-key and the public re-key are consolidated further to generate the re-encryption key.

Abstract: This disclosure relates generally to method and system for time series classification. Conventional methods for time-series classification requires substantial amount of annotated data for classification and label generation. The disclosed method and system are capable of generating accurate labels for time-series data by utilizing a small amount of representative data for each class. In an embodiment, the disclosed method generates a time-series data synthetically and associated labels by using a portion of the representative time-series data in each iteration, and self-correcting the generated labels based on a determination of quality of the generated labels using label quality checker models.

Abstract: Existing approaches for identifying a prosthesis model involve rigorous examinations and visual inspection comparison of X-ray images which is difficult for both radiologists and orthopedic surgeons. This can be a meticulous task that is tedious, dependent on the surgeon's experience, time-consuming and an erroneous recognition can have certain consequences. Method and system disclosed herein provide an approach which involves use of a 3-block classifier for extracting finer features of implant from an X-ray image being processed, and then comparison of the extracted features with manufacturer specifications for identifying manufacturer and type.

Abstract: Hydrogen being a clean, highly abundant and renewable fuel, is a promising alternative for conventional energy sources. Mostly, this hydrogen is stored in the form of hydrides. The existing methods for identification of material for hydrogen storage as expensive and time consuming. A method and system of identification of materials for hydrogen storage has been provided. The method provides a machine learning technique to predict the hydrogen storage capacity of materials, using only the compositional information of the compound. A random forest model used in the work was able to predict the gravimetric hydrogen storage capacities of intermetallic compounds. The method and system is also configured to predict the thermodynamic stability of the intermetallic compound.