Rohitpal Singh Email & Phone Number

Pittsfield, Massachusetts, United States

Greater Boston Area

An Approach for Deliberate Non-Compliance Detection during Opioid Abuse Surveillance by a Wearable Biosensor.Wearable sensors can be used to monitor opioid use and other key behaviors of interest, and to prompt interventions that promote behavioral change. The effectiveness of such systems is threatened by the potential of a subject’s deliberate non-compliance (DNC) to the monitoring. We define deliberate non-compliance as the process of giving one’s device to someone else when surveillance is on-going. The principal aim of this research is to develop an approach to leverage movement and cardiac features from a wearable sensor to detect such deliberate non-compliance by individuals under surveillance for opioid use.• Performed signal processing on 3D accelerometer and physiological signal BVP to remove motion artifacts. Used low pass filter to remove high frequency noises from the target signal. • Extracted 36 and 8 statistical time-domain features from accelerometer and BVP respectively. • Using a personalized weighted random forest classifier (model), we evaluated if a snippet of BVP and accelerometer data received is actually coming from the expected participant or an alternate person. • Analysis of our classifier shows the viability of this approach, as we were able to detect DNC (or compliance) with over 90% accuracy (AUC ROC curve) within 3 seconds of its occurrence.

Methuen, Ma

Natural Language Processing (NLP) and Machine Learning The main motivation is to help users/patients understand their clinical report in terms of whether they have any disease or not. These clinical reports are in the form of unstructured text data. Therefore, text mining helps to discover relevant information in text by transforming the text into data that can be used for further analysis and this is achieved by Natural Language Processing. • Collaborated closely with other team members to plan, design and develop a robust system for determining diseases in EHR.• Researched and performed Natural Language Processing (NLP) on 199 patients' electronic/clinical health records (EHR) to build word features using lemmatization, stemming, bag-of-words, pos tags, report type, section header and tf-idf.• Applied cTakes, an open NLP system for named entity recognition to get annotated semantic disorders from EHR.• Achieved 80% precision by implementing ensemble model (Support vector machine -- Hidden Markov Model) for sequential tagging and classification of words as BIO indicating Beginning, Inside or Outside of disorders.

Greater Boston Area

Education Data Mining. The goal of this research is to increase the effectiveness of education system across USA in real time and to help teachers adopt efficient tools and methods to revolutionize the learning environment using adaptive cognitive assessment. It has few components as follows :• Education data mining with focus on skill decomposition with randomized controlled experiments. Introduced a novel way to determine sub-skills using k-mean clustering with silhouette coefficient.• Performed hypothesis test using t-statistic and p-value (0.05) to determine the significance between templates of each problem set. It indicates importance of relative order of solving a sub skill and this order makes an effective impact on learning of a main skill.• Estimated students' performances through their current knowledge state using conditional probability based Hidden Markov Model .

Noida Area, India

Projects: Xbox Live Ingestion and Data Acquisition Platform (DAP)I worked in Product Intelligence unit at HCL Technologies where I worked on Microsoft Big Data platform "Cosmos" and Data Acquisition Platform (DAP)• Conducted predictive modeling on Azure machine learning platform. Initiated migration plan for 45 workflows.• Deployed big data analysis solutions in production environment of Microsoft’s big data platform Cosmos.• Optimized 18 SQL scripts. Organized and analyzed unstructured data of size 50 TB using Cosmos and scope scripts.• Implemented 25 sophisticated and efficient ETL workflows for large-scale data using SSIS packages and SQL Server.

Master Of Science (M.S.), Data Science, 3.57/4

Bachelor’S Degree, Computer Science And Engineering, 3.4/4

Activities and Societies: • Utsaah Committee • The National Techno-Cultural Festival Committee