Arun is a data science professional with 2+ years of academic research experience in machine learning, data science, computer vision, and artificial intelligence and 1.5 years of industry experience as a DevOps Engineer. Arun is passionate about applying machine learning and deep learning models to real-world applications. Arun has defended his master's thesis, published three research papers and has been awarded Research Assistantship twice during his graduate schooling . Arun's research interests are in the computer vision and deep learning space. Arun has collaborated with various stakeholders in the past and has held responsible positions during his research work. Arun is looking for positions such as Data Scientist, Machine/ Deep Learning Engineer, Computer Vision Engineer, AI Engineer, and Research/ Applied Scientist.
This research work was published at the Artificial Intelligence and Cloud Computing Conference (AICCC 2020). This work performs a thorough investigation of the two types of features used for image classification tasks - learned and hand-crafted features. This work also highlights the significance of convolutional neural networks in image classification tasks. Further, this work shows the significance of CNNs as a feature extractor. This work was a part of the NSF funded project.
GitHub PDF PublicationThis research work was published at The 35th conference on Image and Vision Computing (ICVNZ 2020). This work applies advanced image processing and machine learning techniques in the recognition of physician's gaze. This work uses video interaction and shows the applied methodology could reduce human labor in the annotation process and produces a gaze recognition tool capable of predicting physician gaze with 83% accuracy. This work was a part of the NSF funded project.
GitHub PDF PublicationThis research work was published in the 12th International Conference on Bioinformatics and Computational Biology (BICoB 2020). In this work, the health-span of C elegans in the presence of a life-span extending mutation was explored. Various data analysis, data visualization, video analysis, and statistical techniques were performed in this work. This was a project conducted in collaboration with Rosalind Franklin University.
GitHub PDF PublicationThe goal of this research project is to diagnose age-related macular degeneration (AMD), a common eye disease in the early stages using drusen as biomarkers. This work applies image processing techniques such as segmentation in identifying drusen and uses drusen characteristics and machine learning techniques in identifying the type of AMD.
GitHub PDFThis work automatically predicts the English alphabets spoken by different subjects. The extracted cepstral coefficients are used as attributes and various machine learning techniques such as decision trees, ensemble methods are used in classifying. This machine learning model classifies spoken English alphabets with an accuracy of 93%.
GitHub PDFThis research work applies various image processing techniques on video data and identifies the postures of the nematode C. Elegans. The postures made by the nematode had biological implications on the lifespan and the healthspan of the microbe and accurate identification of postures were required.
GitHub PDFA model that could identify the emotional state of the speaker was built. The emotions identified by the model were angry, sad, happy, fear and neutral. Audio samples from the tamil language were used as attributes for the model and the various prediction algorithms were used to train the model.
GitHub PDFGPA: 4.0
GPA: 3.44
Give me a call or send me an email and I will get back to you as soon as possible!