Abstract: The present disclosure provides systems and methods for performance evaluation of Input/Output (I/O) intensive enterprise applications. Representative workloads may be generated for enterprise applications using synthetic benchmarks that can be used across multiple platforms with different storage systems. I/O traces are captured for an application of interest at low concurrencies and features that affect performance significantly are extracted, fed to a synthetic benchmark and replayed on a target system thereby accurately creating the same behavior of the application. Statistical methods are used to extrapolate the extract features to predict performance at higher concurrency level without generating traces at those concurrency levels. The method does not require deploying the application or database on the target system since performance of system is dependent on access patterns instead of actual data.

Abstract: This disclosure relates generally to data analytics, and more particularly to a system and method for data analytics. The method includes accessing and processing sample data associated with a task to obtain a processed sample data set comprising additional columns than columns in the source data. During processing, metadata corresponding to the additional columns of the processed sample data set is dynamically captured. A plurality of data mining models are iteratively executed using the processed sample data set and the metadata. Said data mining models are obtained by varying variables and versions of a set of data mining models stored in a model repository to obtain a metrics corresponding to the plurality of data mining models. The metrics are indicative of performance of the data mining models. A data mining model is selected from amongst the plurality of data mining models based on the metrics to perform data analytics.

Abstract: Predictive regression models are widely used in different domains such as life sciences, healthcare, pharma etc. and variable selection, is employed as one of the key steps. Variable selection can be performed using random or exhaustive search techniques. Unlike a random approach, the exhaustive search approach, evaluates each possible combination and consequently, is a computationally hard problem, thus limiting its applications. The embodiments of the present disclosure perform i) parallelization and optimization of critical time consuming steps of the technique, Variable Selection and Modeling based on the Prediction (VSMP) ii) its applications for the generation of the best possible predictive models using input dataset (e.g., Blood Brain Barrier Permeation data) and iii) business impact of predictive models that are requires the selection of larger number of variables.

Abstract: Hand segmentation on wearable devices is a challenging computer vision problem with a complex background because of varying illumination conditions, computational capacity of device(s), different skin tone of users from varied race, and presence of skin color background. The present application provides systems and methods for performing, in real time, hand segmentation by pre-processing an input image to improve contrast and removing noise/artifacts. Multi Orientation Matched Filter (MOMF) is implemented and applied on the pre-processed image by rotating the MOMF at various orientations to form an edge image which comprises strong edges and weak edges. Weak edges are further removed using morphological operation. The edge image is then added to the input image (or pre-processed image) to separate different texture region in image. Largest skin-color blob is then extracted which is considered to be correct segmented hand.

Abstract: The present disclosure generates a plurality of time bound test cases based on an Artificial Rain Drop (ARD) algorithm. Here, events associated with an event-based system are compiled in a tabular format. Each of the plurality of events are represented as a regular expression. Further, timed finite automaton is constructed for each regular expression prior to applying the ARD algorithm.

Abstract: This disclosure relates generally to performing user segmentation, and more particularly to predicting attribute values for user segmentation. In one embodiment, the method includes segregating a user with an incomplete attribute value and a user with complete attribute values for an attribute into a first group and a second group respectively, computing prior probability for each suggestive attribute value, identified for the incomplete attribute value, based on number of users in second group having the suggestive attribute value as attribute value for the attribute. Computing likelihood for each suggestive attribute value based on similarity of the attribute values of the user of the first group with users of the second group, computing a posterior probability for each suggestive attribute value based on the prior probability and the likelihood, selecting a suggestive attribute value with the highest posterior probability as the attribute value for the incomplete attribute value of the user.

Abstract: This disclosure relates generally to PPG signal quality assessment, and more particularly to, a system and method for sensor agnostic PPG signal quality assessment using morphological analysis. In one embodiment, a method for PPG signal quality assessment includes obtaining a PPG signal captured using a testing device in real-time, and segmenting into a first plurality of PPG signal samples such that length of each of the first plurality of PPG signal samples more than a threshold length. A signal sufficiency check (SSC) is performed for each first PPG signal sample to obtain at least a first set of PPG signal samples complying with the SSC. A set of features is extracted from the first set of PPG signal samples, based on which each PPG signal sample is identified as one of a noisy and clean signal sample using a plurality of Random Forest (RF) models created during the training phase.