Abstract

The strategic application of artificial intelligence (AI) in healthcare has the potential to revolutionize patient care, particularly in the domains of public health and cardiology. This thesis presents innovative AI-based solutions to address critical challenges in these fields, focusing on enhancing decision-making, disease surveillance, and diagnostic accuracy. In the second chapter, we introduce the AI-Based Home-Based Care Assistive System (AI-HBCAS), developed to aid healthcare professionals in identifying severe patient conditions that require referrals. By analyzing patient demographics, symptoms, and vital signs, AI-HBCAS enhances clinical decision- making, ensuring timely and appropriate care for patients with non-communicable diseases (NCDs). This system aligns with the Ministry of Health and Family Welfare’s goals of improving public health delivery through digital health solutions. The third chapter explores the development of a Pharmaco- surveillance model for outbreak detection and awareness in Primary Health Centers (PHCs). Leveraging AI, we try to analyse prescriptions and pharmacy records to map drug utilization patterns and detect lo- cal outbreaks. The insights generated could potentially support timely public health interventions and contribute to the Ministry of Health and Family Welfare’s objectives of enhancing public health infras- tructure and outcomes. We then delve into the development of a unique deep learning architecture for heart murmur detection using phonocardiogram (PCG) recordings in chapter four. Our modified variable kernel length Resid- ual Networks (ResNets) demonstrate significant improvements in detecting heart murmurs, achieving a weighted accuracy score of 0.708 in the George B. Moody PhysioNet Challenge 2022. This inno- vative approach highlights the potential of deep learning models in improving diagnostic accuracy in cardiology, specifically using audio modalities. Chapter five focuses on the creation of a machine learn- ing approach for predicting outcomes in post-anoxic coma patients using longitudinal EEG and ECG recordings. Our model extracts frequency domain features from these recordings, combining them with demographic data to predict neurological recovery following cardiac arrest. This solution, part of the George B. Moody PhysioNet Challenge 2023, underscores the viability of machine learning models in critical care settings, offering valuable prognostic insights.

Abstract

This thesis explores and analyzes humor generation through several methods. This thesis introduces an approach to automated joke generation that leverages a novel hybrid model combining template extraction and infilling. Recognizing the inherent structure and unpredictability in humor, this work pi- oneers a method that extracts templates from existing jokes and subsequently infills these templates with new content, guided by state-of-the-art (SOTA) techniques available at the time of research. Specifically, the thesis accomplishes three main objectives: Firstly, it develops an automated system for joke genera- tion that innovatively combines template extraction with a deep learning approach for content infilling, ensuring both the preservation of joke structure and the introduction of novelty. This approach uses a rule-based method to extract a template from a joke and then infills it using BERT. Secondly, it extends the base model by incorporating advancements in language models and natural language processing to enhance the flexibility and quality of generated jokes, broadening the scope of humor that the system can produce. This approach uses a novel strategy to identify semantically salient words, and uses LLMs on top of earlier models for the infilling. Lastly, the thesis explores a distinct avenue in humor genera- tion by employing conversational prompts within a conversational AI framework with GPT-4, the SOTA at the time of presentation. This method not only diversifies the ways in which humor can be generated but also reflects on the potential of conversational AI to contribute creatively to the domain of humor. Through rigorous evaluation, this work demonstrates the efficacy of the proposed methods in generating humor that is coherent, contextually appropriate, and, most importantly, funny. This research not only contributes a novel technique to the field of computational humor but also opens up new pathways for future exploration in automated joke generation and the creative capacities of AI.

Abstract

Water, a ubiquitous molecule in life and chemical processes, undergoes autoionization, where it splits into a proton (H+) and a hydroxide ion (OH-). This process plays a crucial role in determining pH and conductivity. However, the mechanism of autoionization remains a mystery due to the vast difference in stability between water and dissociated ions. This difference creates a multiscale kinetic challenge, making it difficult to simulate the entire process using traditional methods. This thesis investigates the water auto-ionization mechanism using well-tempered metadynamics, a powerful tool for exploring free energy landscapes. We employ atomic neural network potentials (NNPs) to overcome the computational limitations of traditional density functional theory (DFT) calculations, allowing us to explore the ionization pathway at a nanosecond timescale. Central to our methodology is the identification of an effective collective variable to drive the ionization in a way that doesn’t break the dynamics of water or disturb the free energy surface. Our approach of using a group of water molecules in vicinity and biasing the molecules to align to a favourable configuration for ionization allows us to initiate ionization on our choice of water molecules. In our thesis we demonstrate the evolution of our collective variable arising from the observed mechanics of auto-ionization along with the various observed patterns in which the ions exist in the system. We have employed a few algorithms to identify the existence of ions. The combination of our collective variables and the analysis algorithms, gives us a more definite molecular understanding of the dynamics defined in previous related work. We demonstrate that our observed dynamics are in excellent agreement with the physical qualities of water and also well represent the dynamics of ionization talked about in previous articles. With the collected data we also attempt to conduct importance classification of various parameters that might describe the initiation conditions of ionization with a good accuracy. In the bigger picture, this understanding may allow us to get closer towards using machine learning neural network potentials to reliably study the dynamics of other water implicit systems and hence gather more information about auto-ionized water and its effects on such systems. In essence this thesis demonstrates the use of neural network potential n2p2 as a reliable method of simulating ions in water without disturbing the dynamics of water or the ions. The use of neural network potential instead of DFT calculations allows us to study simulations with significantly longer timescales within the same processing constraints as before. This can be a significant step towards studying systems that are affected by the auto-ionization of water eg. water implicit protein folding.

Abstract

Sleep staging using Electroencephalogram (EEG) signals is crucial for assessing sleep quality and diagnosing related disorders. This thesis explores methodologies for automated sleep stage classification leveraging EEG data, focusing on both supervised learning and self-supervised learning (mulEEG). The mulEEG study introduces a novel multi-view self-supervised method designed for EEG representation learning. The mulEEG framework utilizes complementary information from multiple views to learn robust representations without requiring labeled data. Key contributions of mulEEG include an EEG augmentation strategy tailored for multi-view self-supervised learning, the introduction of a diverse loss function to enhance complementary information across views, and demonstrating the superiority of mulEEG over traditional supervised methods. The framework achieves this by combining different EEG augmentations such as jittering, flipping, and scaling, and processing these augmentations through both time-series and spectrogram encoders to capture diverse features. This joint training strategy, coupled with a unique contrastive loss function, significantly improves representation learning, making mulEEG highly effective for tasks like sleep staging. Results indicate that the mulEEG method outperforms supervised models in transfer learning scenarios, showing improvements of 1.1% in Cohen’s kappa and 0.85% in accuracy compared to traditional supervised learning models. In a second contribution, we explored supervised learning strategy for EEG sleep stage classification integrating squeeze and excitation (SE) blocks within a residual network and stacked Bidirectional Long Short-Term Memory (Bi-LSTM) units. This model aims to capture complex temporal dependencies and spatial features in EEG data. A significant aspect of this strategy is the application of GradCam for model interpretability, allowing visualization of the model’s decision-making process and allows comparison with sleep experts’ insights. The model was evaluated on several publicly available datasets (SleepEDF-20, SleepEDF-78, and SHHS), achieving high Macro-F1 scores of 82.5, 78.9, and 81.9, respectively. This demonstrates the model’s robustness and applicability across different datasets. Additionally, the study introduces a training efficiency enhancement strategy, reducing training times by eight-fold with minimal impact on performance. In conclusion, these EEG-based sleep staging approaches hold promise for enhancing sleep quality assessment and facilitating the diagnosis of sleep-related disorders, significantly contributing to the field of biomedical informatics. The incorporation of model interpretability provides valuable insights into the underlying decision-making processes, effectively bridging the gap between machine learning models and clinical practice.

Abstract

The simultaneous occurrence of climate extremes poses significant threats to ecosystems, amplifying the vulnerability to various physical risks. A compound extreme event corresponds to a situation in which multiple extremes occur at the same time, leading to more severe impacts than individual extremes. Despite extensive research on individual hot extremes and droughts globally, and the development of numerous indicators to detect and study changes in extreme events across space and time, there remains a substantial gap in understanding the occurrence, magnitude, spatial extent, and associated risks of compound extremes. Most current climate extreme indicators do not provide information on compound or concurrent events. These compound extremes encompass scenarios such as warm/dry, cold/dry, warm/wet, and cold/wet conditions. Comprehensive research integrating multiple climate variables is crucial to address this gap, providing a holistic understanding of the risks and impacts associated with compound extremes. This study aims to fill this gap by investigating various compound extreme scenarios through the analysis of combinations of maximum temperature (Tx) and the Standardized Precipitation Evapotranspiration Index (SPEI). The analysis uses monthly data from 1951 to 2014, obtained from the India Meteorological Department (IMD) for the Indian landmass. The results of the study reveal several key trends and patterns in the spatial extent and frequency of compound extreme events. Notably, the spatial extent of warm/dry events has increased at a rate of 1.8% per decade, indicating a significant upward trend. Conversely, cold/wet events have shown a decrease of 1.1% per decade, highlighting a reduction in these events over the Indian region. Warm/wet events have also exhibited an increasing trend, albeit at a more modest rate of 0.3% per decade, while cold/dry events have shown a modest rise of 0.7% per decade. Furthermore, the study examines the frequency and return periods of these compound extreme events. It is observed that compound warm/dry and cold/wet extremes exhibit high frequency and shorter return periods, posing greater risks to the affected regions. On the other hand, compound cold/dry and warm/wet extremes occur less frequently, indicating longer return periods and consequently lower associated risks. The frequency analysis reveals that across much of the country, the occurrence of warm/dry, cold/dry, warm/wet, and cold/wet extremes ranges from 30-45 months, 15-30 months, 20-30 months, and 25-45 months, respectively. A notable finding of the study is the increased frequency of warm/dry conditions in the recent period (1983-2014) compared to the base period (1951-1982). Specifically, the recent period experienced approximately 31 years of warm/dry conditions exceeding a spatial extent of 5%, whereas the base period had approximately 24 years. This trend underscores the growing prevalence of warm/dry extremes in recent decades. The findings of this study contribute to an enhanced understanding of the changes in compound climate extremes from a multivariate perspective. By examining the interplay between temperature and precipitation extremes, this research provides valuable insights into the evolving nature of climate extremes in India. The observed trends in the spatial extent and frequency of compound extremes have significant implications for climate risk assessment and adaptation strategies. The increasing trend of warm/dry events suggests a heightened risk of heatwaves and droughts, which can have severe impacts on agriculture, water resources, and human health. The study highlights the need for a comprehensive understanding of compound climate extremes to inform effective adaptation and mitigation strategies. Policymakers and stakeholders can leverage the findings of this research to develop targeted measures to enhance resilience against compound extremes. For instance, the increasing frequency of warm/dry conditions calls for improved water management practices, drought-resistant crop varieties, and heatwave preparedness plans. Moreover, the study emphasizes the importance of considering compound extremes in climate models and projections. Traditional approaches that focus on single variables may underestimate the risks associated with simultaneous extreme events. By incorporating compound extremes into climate models, researchers can better inform decision-making processes.

Abstract

In the realm of Natural Language Processing (NLP), enhancing knowledge graphs (KGs) with large language models (LLMs) offers a transformative approach to managing and leveraging structured information. This thesis, titled ”Leveraging Large Language Models for Knowledge Graph Enhancement: From Creation to Utilization,” explores the synergistic integration of LLMs and KGs to enhance various aspects of information extraction, entity classification, and question answering. The research is driven by the recognition of the limitations inherent in traditional KGs, which often suffer from incomplete but dynamic data, and aims to leverage the semantic capabilities of static LMs to address these gaps. The thesis presents a comprehensive study on cross-lingual fact extraction, focusing on extracting factual information from texts in low-resource Indian languages and representing it as English triples. This task is critical for capturing the rich knowledge embedded in these languages, which traditional methods often overlook. An end-to-end generative approach is employed, achieving a significant F1 score of 77.46, thereby setting a new benchmark for cross-lingual fact extraction efforts. This method is validated using the XAlign dataset, which includes multilingual data from Wikipedia articles, ensuring robust evaluation and applicability of the proposed models. Entity type classification is another focal point of this thesis. The research introduces a fine-grained entity type classification model, which is particularly effective in scenarios where KGs are being developed from scratch using only textual descriptions. This model, built upon a multilingual encoder- only language model and incorporating hierarchy-aware losses, utilizes our Wikidata Entity Type Classification (WETC) dataset. By leveraging the structure of Wikidata and the multilingual links in Wikipedia articles, the model achieves significant improvements in entity classification across various languages, attaining a performance of 0.73 Hierarchical F1. The thesis also delves into the integration of KGs with LLMs for question answering (QA). A novel approach, termed Joint Reasoning over KG and LLMs, is proposed, where the structured knowledge from KGs is combined with the generative abilities of LLMs. This hybrid architecture employs graph neural networks (GNNs) and includes modifications such as Prominent Entities for Graph Augmentation (PEGA) and QA-Context-Aware Node Pruning (CANP) to enhance the relevance and accuracy of the QA process.Experimental results demonstrate that this integrated approach significantly outperforms existing methods on various QA benchmarks, including OpenBookQA, CommonsenseQA, and MedQA-USMLE, showing notable advancements in QA accuracy over 20% points. In conclusion, this thesis makes substantial contributions to the field of knowledge graph enhancement using large language models. By addressing the challenges associated with cross-lingual fact extraction,entity classification, and QA, the research paves the way for the development of more advanced and efficient knowledge-based systems by bridging the gap between structured knowledge representation and natural language understanding.

Abstract

The Intergovernmental Panel for Climate Change (IPCC) reported that carbon dioxide (CO2) from anthropogenic emissions is the major contributor to radiative forcing in the atmosphere (IPCC, AR6). The recorded monthly average atmospheric CO2 level in the Mauna Loa Observatory (elevation of about 3397 m) was 421 ppm during March 2023, the highest level since the accurate measurements began 65 years ago. The accelerating CO2 mixing ratios were attributed to changes in land use and land cover (LU/LC). After industrialization (around 1760), CO2 emissions from fossil fuel burning started and became the most significant contributor to the global carbon budget (Ballantyne et al., 2012). Since the pre-industrial period, the burning fossil fuel and LU/LC changes have increased by 40 and 150 % for CO2 and CH4 mixing ratios, respectively (Huang et al., 2016). Recent studies suggest that the CO2 emissions from cement production exceeded global fossil fuel emissions in the past two decades (Andrew, 2018). CO2 mixing ratio levels in the atmosphere are controlled by the combined effect of sources and sinks. Only about half the amount of CO2 released remains in the atmosphere, whereas two major sinks, the terrestrial biosphere, and oceans, absorb the remaining part (Andres et al., 1996). Hence, monitoring and maintaining long-term records of atmospheric CO2 measurements are essential to understanding the carbon cycle and predicting the future behaviour of the controlling factors: photosynthesis, respiration, biomass, fossil fuel burning, and air-sea exchange processes (Machida et al., 2002). The present thesis work is aimed at understanding and assessing the CO2 variability over the Indian region using three approaches: ground-based, satellite data, and model simulations, and further integrating them to provide a comprehensive estimate of CO2 concentrations and their variability. The impact of LU/LC and its changes on CO2 emissions are quantified using ground-based observations and model simulations from 2013 to 2022. In addition, the vertical profiles of atmospheric CO2 were studied to help understand temporal variations in the vertical column, which was used to improve the satellite-based CO2 estimates. A study was conducted to understand the neighbourhood effects on atmospheric CO2 using space-based CO2 data, model simulations, and LU/LC. Further, the objective was to quantify and generate a national CO2 emission dataset as a function of climatic zones, LU/LC, satellite-derived CO2 concentration, and MIROC4-ACTM simulated inverse fluxes.Column-averaged CO2 concentrations (X) were obtained using data from a ground-based Fourier Transform Infrared (FTIR) Spectrometer that was gathered in clear sky days at the NRSC in Shadnagar, India throughout the 2016 period. Calcium Fluoride (CaF2) beam splitter and Indium Antimonide (InSb) detector are used to collect the solar absorption spectra, with the range between 1800 cm-1 to 11000 cm-1 (5.50 μm to 0.90 μm) and the spectral resolution (∆ν) of 0.01 cm-1 for this study. This study analyses spectra using a non-linear least squares Gas Fitting Spectral Analysis (GFIT) developed by the Jet Propulsion Laboratory (JPL), and licenced by California Institute of Technology (Caltech), U.S.A. With the present retrieval scheme, the precision of the FTIR achieved is 0.04 % for XCO2. During the study period, the diurnal amplitude of XCO2 was about 2 ppmv, and the present work compared FTIRretrieved XCO2 data against XCO2 observed by Orbiting Carbon Observatory-2 (OCO-2). A comparative study yields a mean relative bias of -0.78 % and -0.63 % between FTIR data and satellite data for XCO2. The Pearson correlation coefficient (R) between FTIR and satellite data is 0.96 (XCO2, N=13 co-located). In the present work, we also retrieved XCO2 using a groundbased portable EM27/SUN FTIR. The EM27/SUN spectrometers are widely used in the COllaborative Carbon Column Observing Network (COCCON). The advantages of portability of this FTIR allow experiments in a campaign mode so that data gaps in India in terms of geographical coverage can be minimized. The PROFFAST software provided by COCCON has been used to analyse the measured atmospheric solar absorption spectra. This work studied the diurnal variation and the time series of daily averaged XCO2, covering December 2020 to May 2021. The maximum value of XCO2 was observed to be 420.57 ppm. Less diurnal (XCO2~0.44 ppm), but clear seasonal changes are observed during the study period. The FTIRbased XCO2 retrievals help estimate the uncertainties in the satellite-based products and enable the highly accurate products from the satellites by adjusting the biases between ground and space-based retrievals. Thus, ground-based FTIR retrievals are decisive in validating satellite products. An airborne campaign was conducted to investigate the vertical distribution of the atmospheric CO2 using a twin-turbo prope

Abstract

In the field of Quantum Information Theory, the problem of entanglement detection is one of the key areas of research and a long standing one. Entanglement detection problem pertains to finding criteria for a given state to be entangled. One of the prominent methods of detecting entanglement is to find the suitable positive but non-completely positive maps. Positive but not completely positive maps, such as the partial transpose or the Choi map, play a crucial role in identifying entangled states. By applying these maps to a given quantum state, one can derive conditions based on the range of the resulting operators that indicate whether the state is entangled. Various other procedures and criteria have been proposed as well such as Entanglement Witnesses, Reduction Map, CCNR criteria, Range criteria to name a few. Our research work focuses on the positive but non-completely positive maps. We try to develop a generic prescription to construct a positive map and therefore generalise construction of such maps that are useful for the task of entanglement detection. In our work, we study a class of positive maps constructed using the Lindbladian structures. We then show that two of the crucial positive maps i.e transposition and Choi map can be obtained as a special case of a class of positive maps having Lindblad structure. Later, by generalizing the transposition map to a one parameter family we have also used it to detect genuine multipartite entanglement. Additionally, we consider an entanglement measure known as entanglement negativity and we introduce a similar measure for genuine multipartite entanglement.

Abstract

Rivers are significant and low cost source of drinking, irrigation and other water demands across the world. With rising industrial and residential water demands it become essential to manage the water quality and quantity of rivers. The changes in global environmental factors have adversely affected the natural water cycle which in turn has resulted in variation of the natural inflow to river systems. While rapid urbanization, industrialization and population growth had led to rapid decline in river water quality. Without proper management and corrective plan of action for reviving the water quality of rivers there is a risk losing precious natural resources that are essential for continuance of life on earth. This study addresses these challenges within the context of altered environmental conditions that have disrupted the natural water cycle, affecting river inflows and degrading water quality. The management of water quantity in a reservoir-river system possess challenges in terms of forecasting inflows to the reservoir and regulating releases in an adequate manner so that water demands such as irrigation, industries and households can be fulfilled. Accurate predictions of reservoir inflows is critical for water management and is influenced by various climatic phenomenon indices (e.g., Arctic Oscillation, North Atlantic Oscillation, and Southern Oscillation Index) and hydroclimatic variables (e.g., precipitation, evapotranspiration). Recently the use of machine learning techniques for inflow prediction has received a lot of attention however a comprehensive approach that evaluates different algorithms combined with large-scale climate phenomenon indices, and various hydro-climatic variables is missing. To achieve this the present study implements an integrated framework of Machine Learning (ML) algorithms for short-range reservoir inflow forecasting using the climate indices and hydroclimatic data for Bhadra reservoir situated in Karnataka, India. The present study also proposes and evaluates an ensemble model using a robust weighted voting regressor (VR) method to quantify forecasting uncertainty and to improve model performances. The results of the study reveal that the LSTM approach has the greatest influence on prediction accuracy, followed comparably by each model. However, none of the four models used in the study that are Random Forest, GBR, KNN and LSTM seem to be noticeably superior to the VR method. Accessing the water quality of rivers is a challenge and requires the measurement of several water quality parameters. Also the amount of water released from the reservoir has a significant impact on the water quality parameters downstream of the river stretch. An integrated approach for water quality management which takes into account the release of water and the river flow conditions hence becomes important. This study develops and demonstrates the usefulness of such an integrated framework. The developed framework provides a holistic approach to water quantity and quality management by integrating ML algorithms for inflow predictions, numerical methods for reservoir release analysis, Global Environmental Flow Calculator for environmental flow estimations and Qual2k as river water quality model. The developed framework can be easily used to predict and simulate how individual water quality parameters change with changes in water release from the reservoir and flow conditions of the river. In order to ensure that the developed framework can be easily put to use a web application that implements the developed framework using the Bhadra river data is developed. The developed application titled Web-Based Decision Support System for Water Quantity and Quality Management of a Reservoir-River System (WQQM-RR) integrates machine learning algorithms, numerical models, and GIS visualization techniques to predict reservoir inflow, manage reservoir release, estimate downstream water quality, and interactively visualize water quality on a map using GIS. The application uses ReactJS for a responsive and modular frontend, Flask for efficient server-side operations, Open-Street Map for dynamic data visualization, and machine learning algorithms for precise inflow predictions. Additionally, integrates GEFC and Qual2K software for comprehensive water quality estimation. Majorly, the application consists of three major components: inflow prediction, reservoir release management, and water quality estimation with visualization. The inflow prediction module uses historical data on weather patterns, hydrological conditions, and other relevant factors to predict the inflow to the reservoir. The reservoir release management module uses these inflow predictions and storage data provided by the user to manage the release of water from the reservoir. The water quality estimation module uses the Global Environmental Flow Calculator (GEFC) and the QUAL2K water quality m

Abstract

In the natural course of spoken language, individuals often engage in thinking and selfcorrection during speech production. These instances of interruption or correction are commonly referred to as disfluencies. NLP models usually trained on textual data, can have difficulty while working with such data - that is, the audio transcripts from conversational speech. When performing downstream NLP tasks like Machine Translation, we must be aware of the presence of such disfluencies in the training data. These linguistic elements can be systematically removed, or handled as required, to enhance data quality, ultimately improving the output for a given task. For quite some time, disfluencies have been the subject of research. However, most of it is on English, Chinese and other European languages, and not much has been done in this direction on Indian languages. In this thesis, we present a comprehensive research on disfluencies, focussing on the following Indian languages: Hindi, Bengali, Marathi, Telugu, Kannada and Tamil. We carry out some preliminary tasks which help us understand the intricacies of disfluencies in detail - showcasing how it can affect Machine Translation, Questions Answering etc. Initially we studied this on English data, gradually expanding our research focus towards Indian languages. Unavailability of labelled data was the first challenge we encountered while working on disfluency in Indian languages. To address this, we developed some guidelines for annotating disfluencies in Indian Languages. These annotation guidelines is aimed at having a clear and uniform procedure for generating labelled data, thus improving the ‘quality’ of data. These guidelines were used to prepare manually annotated disfluency data. However, developing large amount of data manually for the task is time consuming and needs a lot of effort. To address this labelled-data scarcity problem, we worked on an algorithm to synthetically generate disfluencies. This algorithm aims to facilitate more effective model training for the identification of disfluencies in real-world conversations, thereby contributing to the advancement of disfluency research in Indian languages. The algorithm also attempts to take into account the code-mix nature in Indian Languages - English setting. We combine the manually annotated data with the synthetically generated data for conducting multiple disfluency classification experiments. We trained models for disfluency identification in all the above mentioned languages - having models trained on each language separately as well as varieties of multilingual models.