HKBK COLLEGE of ENGINEERING
(Approved by AICTE & Affiliated to VTU)
22/1, Nagawara, Arabic College Post, Bangalore-45, Karnataka
Email: info@hkbk.edu.in URL: www.hkbk.edu.in
Department of Information Sciences &
Engineering
Prasanna Anthony 1HK19IS015
Hussainkhan AJ 1HK19IS029
Krishan KB 1HK19IS040
Apurva Mahadore 1HK19IS046
TEAM MEMBERS
BATCH NO. 1
Artificial Intelligence/Machine
Learning
Introduction :
• Artificial intelligence generally refers to processes and algorithms that are
able to simulate human intelligence, including mimicking cognitive functions
such as perception, learning and problem solving. Machine learning and
deep learning (DL) are subsets of AI.
• Specific practical applications of AI include modern web search engines,
personal assistant programs that understand spoken language, self-driving
vehicles and recommendation engines, such as those used by Spotify and
Netflix.
Limitations:
• Time complexity
• Space complexity
• Heavy computation power
Creative Director AI
• Creative AI is a new branch of artificial intelligence in which AI can create paintings,
write compelling stories and compose new music.
• We are planning to use Deep Learning and Neural Network model and DeepDream
algorithm.
• One of the state-of-the-art methods is called Deep Dream.
• By reading the variations of a few sample photos as input and combining them, the Deep
Dream algorithm can produce an artistic work. More bizarre new traits would start to
appear as the network model received more and more photographs.
• The same principle would apply to music and stories, which would read a few input files
and produce an aesthetic work. We would use this technique to give pictures and videos
a surreal appearance.We are planning to build a Creative AI model which would enhance
the arts field in the artistic form of images, music, and stories as the final destination of
our project.
• We are using the DEEP DREAM Algorithm for artistic image creation and this could be
developed in a technological field of art and also beneficial form in creating NFT.
Stock Predictor AI
• •The question is, can we make machines predict the value of a stock? Scientists, analysts,
and researchers all over the world have been trying to devise a way to answer these
questions for a long time now.
• Our approach will be to use time series prediction, which will analyze specific individual
stocks over time and predict their future performance.
• All business presentations, quarterly publications, and speeches made by representatives
will be analyzed using NLP. Further, we plan to integrate a web crawler.
• •Sentiment analysis of a particular correspondence.
• •Our model would search social media sites for tweets or news that were related to the share
we wanted to predict. This would categorize the correspondence under the categories of
sentimental value and alter the value of future predictions.
Healthcare
• Deep learning has been proven to be superior in detecting disease from X-rays, MRI scans,
and C.T scans, which would significantly improve the speed and accuracy of detection.
• We have been wanting to improve the speed and accuracy of detecting and
localizing tumors based on MRI scans. This would drastically reduce the cost of cancer
diagnosis and help in the early diagnosis of tumors, which would essentially be a lifesaver.
We would give MRI scans as input to our developed model that could localize tumors in
the specific area of the body.
• We will train and develop a few models. The initial test is designed to detect whether a
tumor exists in the brain MRI scans. The image will be fed to a segmentation rest until the
model locates and segments the tumor.
• We would use transfer learning in this project to speed up the process of optimization.
• We would evaluate trained to resonate classifier network and Risk Unit segmentation
Network on testing data. Apply APIs to build deep convolutional neural networks.