Data Science - Artificial Intelligence - Data Architectures
Portfolio
About
I am Artificial Intelligence enthusiast and develop intelligent products, to address real life problems and open challenges. My current areas of research are Big Data and Deep Learning algorithms for Computer Vision, Audio Processing, Natural Language processing and Self driving Cars.
As a part of my Job, I have been working with Big Data Technologies and developing scalable Data Mining and Artificial Intelligence platforms using Distributed computing and big data paradigms while adopting data lake practices using Hadoop, Spark, Kaka, NiFi, NoSQL, Akka etc.
I am writing a robotic scrum master that can conduct scrum meetings and keep a track of everyones progress and manage backlogs. Scrum meeting is an essential component of Agile Software Development Methodologies, where a person who plays the role of scrum master has to keep a check on the progress as per the sprint planning.
Rather than just creating a piece of software that we can interact through GUI, I am aiming to create voice interactive agent which can generate minutes of the meeting, take inputs from the developers about their tasks and track the same for the next sprints.
Once I will achieve this much, I would like solve some optimization problems in Agile SDLC, related to sprint planning based on Business Value related to each story and the dependencies between them.
Automatic music classification/recognition is one such area which is being widely used in many commercial applications also like Shzam, Google Play, Sony Track ID, etc. All these applications have one thing in common that they aim to understand the semantic of the music rather than just curating the metadata out of it. To develop an advanced and intelligent music player there is a large semantic gap between audio signal processing and listeners” preference. Most of the cloud based music providers use collaborative filtering and sound Meta data to recommend the next song to the listeners. But they could not fulfil the gap of listener’s preference i.e genre, mood, lyrics, instrumentation, rhythm, music records time etc.
The problem of automatic text summarization is one that has garnered significant interest in recent years. Humans want to read a short gist of a news article before deciding whether or not they want to spend time reading the main article. Hence, there is a need to build computers that can read a piece of text and give a short summary. Automatic document summarization can be done in two ways. We have chosen extractive summarization, which identifies important words/phrases from the source document and creates the summary. Headline generation is a special case of text summarization, which builds a concise headline for a document.
In this project the data was scrapped from various webistes and the corresponding headlines were generated.
Each recipe consists of:
A recipe title
A list of ingredients
Preparation instructions
An image of the prepared recipe (missing for ~40% of recipes collected)
The model was fitted on the recipe ingredients, instructions and title. Ingredients were concatenated in their original order to the instructions. Recipe images were not used for this model.
Generated: Asparagus with Chicken
Original: Asparagus and Dill Avgolemono Soup
Recipe: asparagus ; chicken stock ; unsalted butter ; leek ; onion ; ribs celery ; salt ; water ; eggs ; juice of 2 lemons ; minced fresh dill ; dill sprigs for garnish ;Trim off ends of asparagus and using a vegetable peeler remove about 3 to 4-inches of the skin of each stalk , reserving both the ends and peels . Cut asparagus into 1-inch pieces , reserving tips for garnish . In a saucepan combine the asparagus peels and trimmings with the chicken stock , bring to a boil , remove from heat and allow stock to infuse for 15 minutes . Strain stock and reserve . In a pot of salted boiling water blanch the asparagus tips for 2 to 3 minutes , or until brilliant green and barely tender , and then refresh in a bowl of ice water . When tips are chilled , drain and reserve . In a large heavy pot melt the butter over moderate heat and cook the leeks , onion and celery , seasoned with salt and pepper , until softened , about 5 to 8 minutes . Add the 1-inch asparagus pieces and stir to combine ...
Neural machine translator for English2German translation.
This project is about translation from English to German language. That is accomplished by using TensorFlow library, more specifically it utilizes embedding with attention sequence-to-sequence model.
The goals / steps of this project are the following:
Perform a Histogram of Oriented Gradients (HOG) feature extraction on a labeled training set of images and train a classifier Linear SVM classifier.
Optionally, you can also apply a color transform and append binned color features, as well as histograms of color, to your HOG feature vector.
Note: for those first two steps don't forget to normalize your features and randomize a selection for training and testing.
Implement a sliding-window technique and use your trained classifier to search for vehicles in images.
Run your pipeline on a video stream (start with the test_video.mp4 and later implement on full project_video.mp4) and create a heat map of recurring detections frame by frame to reject outliers and follow detected vehicles.
