Using NLP, convert unstructured data to structured format to get the useful information in terms of Entity. Domain of data related to attack information. Where I have to extract the info to whom killed or injured, count and when etc.

Using artificial intelligence which conducts a conversation via auditory or textual methods. Such programs are often designed to convincingly simulate how a human would behave as a conversational partner, thereby passing the Turing test. Team has created the front end pipeline to communicate with services.

4 years old data is used for survival analysis, which gave 92?curacy on predicting the life of a component in a running machine. This saved a huge cost in replacing the component before they meet the end of life, and availability of components before them going out of service.

Detection of food Items in raw and frozen stage in real time, using the Deep Convolution Neural Networks . Optimization of Deep Learning models to make them in size of 18MB for 40 different types of food categories, with an accuracy of 97%.

Including the identification of percentage age of cooking of raw food to final cooked food, Providing the cooking assistance to the cooks for entire cooking sessions by extracting the important features using Image Processing (OpenCV) algorithms and Machine Learning (SVM) based models.

Trained the NextGenMFP engine on 10K resumes for 100 different JDs, and able to predict the suitable jobs and to a input resume from the resume database, without the external assistance (or human resource manager). It was 97?curate system, which used the Machine Learning concepts like ARM, SVM, K-Means, Venn Diagram

Abstract: Designed a system to retrieve different resumes for different employers from the same set of resumes depending on the weightage given by the employer to four different fields that are Domain, Location, Education, Qualification and Experience.

The main purpose of the project “People Relation Map” is to build a system using machine learning capabilities to analyse over the Microsoft Outlook E-mails exchanged. This analysis will give the overview in form of Strength calculator, and Adjacency Matrix between different e-mail users.

Technologies used: RF Model, Feature Selection Method such as RFE, Email Processor, SMTP extractor, Footer Remover of the Legal as well as Quoted text information using SVM model, Strength Calculator

Objective: Reduce the time and effort in searching the contact person history of contacts, so the delegation of work will go smoothly

The main purpose of the project “Auto Inspection” is to build a system using machine learning capabilities to auto inspect the print defects with in-line scanner unit without human intervention. Auto inspect of print defects will be done by comparing the reference images and scanned images [Object Images] on the printed media. The reference image used for defect identification is Proven Image.

The main purpose of the project “Failure Prediction” is to identify the probability of the machine failure after a particular time period and also to find the components which will impact the cause of failure.