In this project, the goal is to build a pneumonia detection system, assisting physicians to make better clinical decisions or even replace human judgement in certain functional areas of healthcare. Aim is to locate the position of in ammation in an image. Handling the data with generators and using the transfer learning technique to build the Mask RCNN and ResNet Neural Network model solved the problem. Used techniques of Computer Vision to locate the position of the in ammation. The developed model was about 94.8% percent accuracy score,85% mean_iou, and a loss of 10%.

The objective of this project is to build a face recognition system, which detects and recognizes faces. Along with an interactive system that interacts by dialogues accordingly. Used techniques of Siamese Networks and Keras for recognition of the face, CNN to locate the position of the face. Used Text to Speech for and NLP algorithms to make the program talk and convert the user input voice to text. The developed model was about 96.07% accuracy score.

Objective of this project is to build a face recognition system, which includes building a face detector to locate the position of a face in an image and a face identification model to recognize whose face it is by matching it to the existing database of faces. '

• Used techniques of Computer Vision, Keras and CNN to locate the position of the face. Siamese Networks for recognition of the face.

• The developed model was about 96.07?curacy score.

In this project, the goal is to build a pneumonia detection system, assisting physicians to
make better clinical decisions or even replace human judgment in certain functional areas of
healthcare.
 Aim is to locate the position of inflammation in an image. By taking CXR image of
patient as input and predict the potentially infected-area by positioning the bounding box .
 Problem faced here is to handle and train model with large imbalanced data and fine tune it.
As this project is of medical use making the model is to be accurate as possible is another
problem faced.
 Handling the data with generators and using the transfer learning technique to build the
Mask RCNN and ResNet Neural Network model solved the problem. Used techniques of
Computer Vision to locate the position of the inflammation.
 The developed model was about 94.8% percent accuracy score, 85% mean_iou and a
loss of 10%

Objective of this project is to predict the likelihood of a liability customer buying personal
loans. Identified potential loan customers for Thera Bank using classification techniques.
 Obstacles faced here are to get the data required, understand them and the the terms and
conditions. Understanding the relationship between the attributes and best suitable model
was one of the major problem.
 Used techniques of Logistic Regression and KNN algorithm in order to select the best
performing one.
 Used the concept of precision, recall and ROC curve to identify the best working model.
 The developed model was about 96.07?curacy score.