Ongoing Research
Check our Ongoing Research Projects
Research Led by Dr. Asmaa Elbadrawy
Dr Asmaa and her students explore datasets, data analysis methods, and applications related to sustainability. Sustainable living is becoming more and more crucial over time, especially with the growing concern about drained earth resources. Sustainability can be practiced on a personal level, family level, community level, or within a certain industry. Topics such as “Sustainable Living”, “Zero-Waste Lifestyle”, and “Zero-Waste Manufacturing” are all related to sustainability.
The project has an exploration component, where we explore datasets, searching for interesting patterns that may be of interest to communities concerned with sustainability. The project also has an application component, where we potentially develop applications that involve data analytics methods that can help various communities apply sustainability practices within various contexts.
Analysing Oil and Chemical Spill Incidents and Clustering Spill incidents in US shores: An Applied Project by IFT Master’s students Sudeep Joel Maguluri, Muni Sandeep Kumar Ravilla and Anudeep Gottipalli- Fall 2023
The project aims to extract valuable insights from a spill incidents dataset that spans a significant timeframe. Through advanced data analysis and machine learning techniques, the project seeks to uncover trends and patterns. Key phases include data preprocessing, geospatial extraction, and text augmentation using NLTK and GPT-3. Text classification, addressing class imbalance with SMOTE, and visualization techniques are employed to provide stakeholders with insights for informed decision-making regarding spill incidents.
Integrated Air Quality Monitoring System: An Applied Project by IFT Master’s students Chitresh Jain, Nisha Pacharne and Pratik Agarwal- Fall 2023
This project establishes a comprehensive air quality monitoring system integrating both static dataset analysis and real-time data collection through an IoT device. The static dataset, inclusive of AQI values and pollutant categories, undergoes analysis on Amazon Web Services (AWS) and visualization on Tableau. Concurrently, an IoT device, employing the ESP8266 microcontroller and environmental sensors, collects real-time data on diverse air quality parameters. This IoT-generated dataset is analysed and visualized on the ThingSpeak platform, offering dynamic insights through graphical representations. The project envisions future integration of email notifications for threshold breaches. This can provide timely alerts when air quality parameters surpass predefined thresholds, facilitating swift responses to environmental concerns and contributing to heightened environmental awareness.
Machine Learning Analysis on the AWS Cloud: Impact of Sustainable Energy on GDP – An Applied Project by IFT Master’s student Anup Bhatt- Fall 2023
The project aims to explore the relationship between the adoption of sustainable energy practices and their potential impact on a country’s Gross Domestic Product (GDP). This project involves leveraging machine learning techniques and tools hosted on the Amazon Web Services (AWS) cloud platform for data analysis.
Phoenix Summer Climate Trends and their Impact on Humans: An Applied Project by IFT Master’s students Akshata Arun Parulekar, Ajeet Giri and Nikitha Kakuru- Fall 2023
The “Phoenix Summer Climate Trends and their Impact on Humans” project aims to develop an interactive dashboard to illustrate the correlation between climatic trends and human health in Phoenix, Arizona, from 1950 to 2023. It focuses on analyzing historical climate data, particularly during summer months, to understand extreme weather’s influence on the local population and mortality rates. By comparing urban and rural climate conditions, the project seeks to uncover disparities and inform public health strategies. Rooted in data-driven analysis, the initiative aims to enhance decision-making and foster resilient planning to address evolving climatic conditions, emphasizing comprehensive methodologies and user-friendly visualization techniques for stakeholders.
SustainabilityBot – Sustainability News Research Tool: An Applied Project by IFT Master’s student Pravesh Ponnuvelu- Fall 2023
SustainabilityBot is a news research tool designed to facilitate effortless information retrieval from sustainability-related articles and news sources. Leveraging state-of-the-art Natural Language Processing (NLP) techniques, the tool integrates OpenAI’s GPT-3.5 Language Model (LLM) for semantic understanding and FAISS for efficient similarity searches. Users can input article URLs or upload text files containing URLs, fetching article content for analysis. SustainabilityBot generates embedding vectors for articles, constructs a FAISS index, and provides users with real-time answers to sustainability-related queries.
Eco-Ed: A website published as part of an Applied Project by IFT Master’s students Abarna Asokan and Swetha Shree Byllahali Ananthaswamy– Spring 2023
The purpose of Eco-Ed is to improve awareness among students on the local community’s solid waste management, energy consumption, water consumption and management by creating an interactive UI with visualization on three main topics of sustainability – landfills, waste water, energy consumption.
Mapping Food Insecurity: A Dashboard Published as part of the Applied Project by IFT Master’s students Smruti Govindrajan and Parinita Vedantam– Spring 2023
Our Tableau dashboard presents various visualizations to provide insights into global food insecurity.
Phoenix Sustainability Dashboard: A Dashboard published as part of the Applied Project by IFT Master’s students Achint Bhat, Gwendolyn Holiness– Spring 2023
The Phoenix Sustainability Dashboard is a comprehensive tool designed to help businesses and organizations in the Phoenix metropolitan area track and monitor their sustainability performance.
https://public.tableau.com/app/profile/achint.bhat/viz/WasteDisposalDashboard/Dashboard1
Testing My Waters: A Dashboard published as part of the Applied Project by IFT Master’s students Anushree Misra, Yash M. Pachchigar– Spring 2023
Allowing users to interact with a dashboard detailing water use, scarcity, and quality for the state of Arizona.
https://public.tableau.com/app/profile/anushree.misra/viz/TestingMyWaters-Arizona/Main
Research Led by Dr. Brian Atkinson
Students: Swaroop Damodaran, Jaison Davis
Sarcasm is a common form of expression in everyday language, but it can be difficult to detect in text without the use of context or tone of voice. Natural Language Processing (NLP) techniques have been developed to address this issue by analyzing the linguistic features of text to identify patterns of sarcasm. These techniques involve training machine learning algorithms to recognize certain words or phrases that are commonly associated with sarcasm, as well as looking at contextual cues such as negation or exaggeration. While sarcasm detection is still a relatively new area of research, NLP has shown promising results in identifying sarcasm in online conversations, social media posts, and even in written works of literature. As technology continues to advance, it is likely that NLP will become an increasingly important tool for identifying and understanding the complexities of sarcasm in text.
Research Led by Dr. Tatiana Walsh
Integrated Attendance and Identity Verification System
The project aims to develop an integrated solution for class attendance using Face Recognition with Email Authentication. Beyond traditional face detection, the learning objectives cover understanding face recognition algorithms, preprocessing data for model training, building the model using OpenCV and face recognition libraries, and integration into an attendance system. This innovative project automates attendance processes, eliminating manual efforts.
In addition to facial recognition, the project introduces email authentication to enhance student verification during classes and special events like exams. During such events, students receive a One-Time Password (OTP) via email, ensuring a secure and efficient means of identity verification. This dual authentication method provides a faster, hardware-flexible alternative and addresses privacy concerns with an end-to-end encrypted email authentication process.
The project underscores the multifaceted benefits of this integrated approach, including speed, hardware simplicity, reduced impersonation risk on social platforms, bot mitigation, and a privacy-focused stance. It is a comprehensive solution, ensuring accurate class attendance and robust identity verification during critical academic events.