Abstract

Cuff-less blood pressure (BP) is essential for continuous health monitoring to prevent diseases such as hypertension. Due to the discomfort caused by inflation and deflation of the cuff, it is not possible to monitor continuously. Although Pulse Arrival Time (PAT) derived from Electrocardiogram (ECG) and Photoplethysmogram (PPG) for cuff-less Blood Pressure (BP) measurement has been a contemporary and widely accepted technique. However, the features extracted for it are conventionally from an isolated pulse of ECG and PPG signals. As a result, the estimated BP is intermittent. Multi-parameter models are developed using regression analysis, to estimate systolic blood pressure (SBP) and diastolic blood pressure (DBP). Hence, the correlation of multiple extracted features with the blood pressure is proved. To achieve this, simultaneous electrocardiogram (ECG) and photoplethysmographic (PPG) along with respective BP data were collected for 171 patients as per Association for the Advancement of Medical Instrumentation (AAMI) standards from Care Hospital, Hyderabad. During data collection, the simultaneous ECG and PPG data was collected from (Food and Drug Administration) FDA approved Vios Medical System (VMS) and BP data was collected using mercury based sphygmomanometer, manually after the required training. The developed algorithm uses wavelet transformation on ECG and PPG signals for detection of the occurrence of essential wave points precisely, even in the presence of artifacts. The feature vector of dimension 32, has been created using the fiducial points detected from ECG and PPG signals. The SBP and DBP estimation model has been obtained using regression analysis. The proposed estimation model provides the error within the range proposed within AAMI and British Hypertension Society(BHS) standards for BP estimation. In addition to this, Central BP (CBP), which is more closely associated with the cardiac activities, has also been estimated. Data for 33 patients was collected from Cath lab and CBP estimation model has been trained using PLSR regression analysis. As a resultant, fair amount of estimation accuracy has been achieved. Further, hardware design has been proposed which will simultaneously measure ECG and PPG signal and transmit the data to processing unit via bluetooth. The data received at processing unit will be processed and peripheral BP and CBP are estimated using the proposed algorithms. This continuous real-time BP monitoring technique can be useful in the treatment of hypertensive and potential-hypertensive subjects

Abstract

The pervasive nature of music in human lives, from day to day activities to large scale concerts makes it an important part of human social interactions. From an evolutionary perspective, it has been argued that music has the capacity to evoke shared responses within groups, thereby allowing it to act as a form of social glue increasing group cohesion. One proposed mechanism for these shared responses is empathy. Over the last decade, a great deal of research has sought to identify the behavioural mechanisms underlying the affective capacity that music affords. These studies have put forth various theories that suggest an integral role played by empathy in inducing emotions while listening to music. In this thesis, we try to expand our understanding of empathy in the context of music processing from both behavioural and neural fronts. Through an extensive literature review, we find that previous studies on empathy have sought to examine it in terms of two distinct but related subsystems: The cognitive subsystem, which is concerned with top-down aspects of empathy such as motor-mimicry and the affective subsystem, which primarily deals with bottom-up aspects related to affect processing such as emotional contagion. In the first study, we looked at behavioural differences in self-reported continuous ratings of ”feeling moved” while participants listened to continuous pieces of music. Using a data-driven approach to cluster the ratings, we found significant differences in ”feeling moved” to both sad and joyful music modulated by participants’ affective empathy. Our results provided empirical support to the aforementioned Cognitive vs. Affective framework of empathy from a behavioural standpoint. In the second study, we sought to extend our earlier work by examining how empathy modulated brain responses while listening to music using Functional Magnetic Resonance Imaging (fMRI). Prior to the current study, neuroimaging studies investigating the role of empathy in music processing have been scarce and did not examine functional connectivity in the brain. Our study overcomes this limitation by examining both global and localised processing of music modulated by empathy through graph-theory based methods. Overall, we found differences in global and local connectivity patterns in music processing modulated by participants’ trait empathy. Participants scoring higher on cognitive empathy tended to recruit mechanisms related to motor mimicry, whereas those scoring higher on affective empathy were found to recruit regions related to auditory affect processing. Both of our studies corroborate previous research in the field in addition to providing novel findings and as a result further supporting the existence of two subsystems in Cognitive vs. Affective framework of understanding empathy.

Abstract

Management of traffic is a very challenging task and it gets even more challenging when we take into consideration real-world problems like urban pollution and emergency evacuation. The standard traffic management approaches aim to solve travel demand forecasting which is essential for long-term planning and evaluation of transport policies of a region, but all these approaches propose only one solution. Whereas, in this thesis we propose multiple optimal solutions maintaining the key characteristics across all the solutions known as Pareto optimal solutions to be used in different situations as suitable. Having alternate solutions to route vehicles during events such as emergency evacuation and pollution control can be quite useful and we demonstrate with our results that they solve the problem in a much better way. Rising levels of pollution is a major concern across many parts of the world and has been widely recognized for its adverse impact on human health. To address this, we develop a transportation policy in this thesis to handle air pollution caused by the heavy flow of traffic in urban areas. In particular, we aim to distribute the traffic flow more evenly through a city, by developing a flow algorithm that computes multiple solutions, each of which accommodates the maximum flow. To tackle this, we develop a Paretooptimal Max Flow Algorithm (PMFA) to suggest multiple max flow solutions and introduces the notion of k-optimality into PMFA to ensure that the suggested Pareto solutions are sufficiently distinct from each other – referred to as Pareto-k-optimal Max Flow Algorithm (k-PMFA). Events during an emergency unfold in an unpredictable fashion which makes management of traffic during emergencies pretty challenging. Furthermore, some vehicles would need to be evacuated faster than others e.g. emergency vehicles or large vehicles carrying a lot more people. The Prioritized Routing Assistant for Flow of Traffic (PRAFT) enables prioritized routing during emergencies. However, the PRAFT solution does not compute multiple plans that can help to handle better the dynamic nature of emergencies. PRAFT maps the prioritized routing problem to the Minimum-Cost Maximum-Flow (MCMF) problem, hence its solution can accommodate maximum flow while routing vehicles based on priority (maps higher priority vehicles to better quality routes (i.e., ones with minimum cost)). We build upon the PRAFT solution to develop a Pareto Minimum-Cost Maximum-Flow (Pareto-MCMF) algorithm which can compute all the possible MCMF solutions. To summarize, we study the following two traffic management problems in this thesis, 1. To better manage traffic for urban pollution control: (a) We develop a Pareto-optimal Max Flow Algorithm (PMFA) to suggest multiple max flow solutions. (b) Introduced the notion of k-optimality into PMFA to ensure that the suggested pareto solutions are sufficiently distinct from each other – referred to as Pareto-k-optimal Max Flow Algorithm (k-PMFA). 2. To better manage traffic during emergency evacuation: (a) We develop a Pareto Minimum-Cost Maximum-Flow (Pareto-MCMF) algorithm which can compute all the possible MCMF solutions. Through a series of experiments performed using the well-known traffic simulator SUMO (a) By performing emission modelling on the New York map, we show that our policy distributes the air pollution more uniformly across locations (b) We show that all the solutions generated by Pareto-MCMF indeed have properties similar to a MCMF solution thus providing multiple high-quality options for traffic police to pick from depending on the situation.

Abstract

Today the communication technology relies on state of the art in networks, which is lead by 5G. Polar codes are one such class of codes known to achieve very high speed in 5G. Any message that needs to be transferred from a noisy channel requires encoding and decoding to keep the message intact and not let the noise corrupt it. Using polar codes, transmission can be done at a rate arbitrarily close to capacity and decoded at the receiver with very small probability of error. These codes, introduced by Arikan, achieve the capacity of arbitrary binary-input discrete memoryless channel W under successive cancellation decoding. For any such channel having capacity I(W) and for any coding scheme allowing transmission at rate R, the scaling exponent is a parameter that characterizes how fast the gap to capacity decreases as a function of code length N for a fixed probability of error. Scaling exponent for small-sized kernels up to L = 8 has been exhaustively found, but as kernel size increases, the complexity to find this exponent increases. In this thesis, we find a simple solution to reduce complexity of finding the scaling exponent for a specific type of kernel. We consider product kernels TL obtained by taking Kronecker product of component kernels and derive the properties of polarizing product kernels relating to the number of product kernels, self-duality and partial distances in terms of the respective properties of the smaller component kernels. Subsequently, the polarization behavior of component kernel Tl is used to calculate scaling exponent of TL = T2 ⊗Tl , which has lower complexity as compared to the basic method. Using this method, we show that µ(T2 ⊗ T5) = 3.942. Further, we employ a heuristic approach to construct a good kernel of L = 14 from kernel having size l = 8 having best µ and find µ(T2 ⊗ T7) = 3.485. Plotting all these values, we make some interesting observations about the nature of this parameter as kernel size increases. Further, the work of this thesis which focuses on finding scaling exponent for a specific type of kernel (T2 ⊗ Tl), can be extended to any general kernel. This will result in calculating scaling exponent much efficiently. Another interesting problem to solve is to find if the scaling exponent for modified polar codes like polar subcodes or RM codes exists and, if so, finding those values. RM codes are known to polarize as well, and an algorithm to find the basis matrix of any RM(m, r) codes are presented in the thesis, which leads to the generator matrix of these codes needed for encoding. Another category of codes that are very similar to RM codes are twin codes, and it is conjectured that under certain conditions, twin codes are equal to RM codes for binary symmetric channel (BSC). We give a detailed proof of them being same till m = 4 and conclude the proof for a particular case of any general m.

Abstract

Landscapes are the composition of dynamic components of complex ecological, economic, and cultural elements on which human and other life forms depend directly. Landscape dynamics driven by land use land cover (LULC) changes due to anthropogenic activities are affecting ecology, biodiversity, hydrological regime, and hence people’s livelihood. There has been increasing apprehensions about environmental degradation, depletion of natural resources due to uncontrolled anthropogenic activities, and its consequences on long-term sustainability of socio-economic systems around the world. This necessitates an understanding of landscape dynamics and the visualization of likely changes for evolving appropriate strategies for prudent management of natural resources. Modeling of forest cover changes offers to incorporate human decision making on land use in a systematic and spatially explicit way through an accumulation of land use choices, social interaction, and adaptation at various levels. Several models developed by the research community so far has largely been utilized to evaluate the empirical studies, explore theoretical aspects of particular systems rather than forecasting their effectiveness across the various landscapes representing bio-physical dissimilarities. Hence, there is a need to demonstrate an appropriate modeling technique, that captures the current degradation in an effective way as compared with the traditional agent-based or non-agent based land use change modeling techniques. In this regard, the objectives of current research are to understand and model the spatiotemporal patterns of landscape dynamics in the Uttara Kannada district of Central Western Ghats. This involves, (i) developing an appropriate modeling framework incorporating the spatiotemporal changes in the landscape at the regional level; (ii) implementing a hybrid model to capture the changes at the landscape level by integrating bio-ecological aspects with socio-economic growth; (iii) evaluating the environmental conditions in response to scenarios of drivers of change like developmental policies and their potential impacts; (iv) assessing the likely scenario of the landscape dynamics based on conservation policies of ecologically sensitive regions (ESR) and other recommendations. The vegetation dynamics quantified using spatial data acquired through spaceborne sensors along with collateral data shows a decline in vegetation cover from 92.87% (1973) to 80.42% (2016). Land use analyses through supervised classifiers based on the Gaussian maximum likelihood algorithm reveals a deforestation trend as evident from the decline of evergreensemi evergreen forest cover to 29.5% (2016) from 67.73% (1973). In addition, agricultural spatial extent (7.00 to 14.3 %) and the area under human habitations (0.38% to 4.97%) have also shown a steep increase. This has also led to forest fragmentation (interior forest cover lost by 64.42 to 22.25 %) in the district. In order to visualize the likely changes, the current work proposes a modified Hybrid Fuzzy-Analytical Hierarchical Process-Markov Cellular Automata model by accounting for the land use changes and to evaluate the role of policy decisions. The impacts are noticed at the microscale (landscape level) with policies that are framed at a macro level and how they propagate under various scenarios. To understand the landscape dynamics in the region, modeling has been carried out under four scenarios to account for potential changes driven by economic growth and climatic aspects at a landscape level. Modeling and visualization further confirm the loss of forest cover in near future with an increase in monoculture plantations from 14.8 to 17.97% and an increase in built-up area from 4.81 to 9.30 % by 2022 under business as usual (BAU) scenario. The proposed hybrid modeling approach with the constraints in the cellular automata technique has been used to simulate various scenarios (i) managed growth rate (2022), (ii) IPCC climate change rapid growth (2031, 2046), (iii) policy-induced constrained Ecological Sensitive Regions. The rapid growth rate scenario highlights a likely loss of forest cover by 11.1%, with an increase in plantations covering 20.9% and built-up as 10.2% of the region by 2046. Land use changes assessed through considering constraints of Ecological Sensitive Regions (ESR1) and the protection of intact or contiguous (interior) forest patches, highlights the role of policy decisions in land use changes. ESR-1 protection scenario shows forest cover is likely to remain at 48% (2021) and 45% (2031) though there is an increase in built-up area from 5.8 to 7% (2031) and agriculture area. The comparison of policy scenario-1 (ESR-1) and scenario-2 (protection of interior forest) depicts scenario-1 focuses more on conservation and limits the growth to the ESR- 2, 3 and 4 regions, whereas scenario-2 shows growth can occur throughout the

Abstract

Identifying language information from speech utterance is referred to as spoken language identification. Language Identification (LID) is essential in multilingual speech systems. There are various conditions under which the performance of LID systems are sub-optimal, such as short duration, background noise, channel variation, and so on. There have been efforts to improve performance under these conditions, but the impact of speaker emotion variation on the performance of LID systems has not been studied. Hence in contrast to the previous studies, for the first time in the literature, the present work investigated the impact of emotional speech on language identification. In this work, different emotional speech databases have been pooled to create the experimental setup. A dataset of this kind was not available for LID, and is a contribution of this thesis. Additionally, state-of-art i-vectors, time-delay neural networks (TDNN), long short term memory (LSTM), and deep neural network (DNN) x-vector systems have been considered to build the LID systems. Performance of the LID system has been evaluated for speech utterances of different emotions in terms of equal error rate and Cavg. The results of the study indicate that the speech utterances of anger and happy emotions degrades performance of LID systems more compared to the neutral and sad emotions. To that effect, we investigated adaptation approaches for improving the performance of LID systems by incorporating emotional utterances in the form of adaptation dataset. Hence, we studied a prosody modification technique called Flexible Analysis Synthesis Tool (FAST) to vary the emotional characteristics of an utterance in order to improve the performance, but the results were inconsistent and not satisfactory. Therefore, we propose a combination of Recurrent Convolutional Neural Network (RCNN) based architecture with multi stage training methodology, which outperformed state-of-art LID systems. Keywords: TDNN, LSTM, DNN x-vector, Flexible Analysis Synthesis Tool, Recurrent Convolutional Neutral Network, multistage training

Abstract

Last decade has witnessed a growing role played by online platforms in enabling socio-political movements and shaping narratives. Social networking sites have evolved into being active sites of political culture and begin to function as a “public sphere”. This has also enabled users on these platforms to conduct campaigns and coordinate acts; engaging in collective user practice of “trolling” or shutting down voices. We observed a recurring form of online collective action that played out on Indian online platforms. Users organise over Twitter to initiate a collective uninstalling/down voting of an app endorsed by a celebrity they have ideological disagreements with. The logic is simple: it is a call to boycott the product the associated celebrity is endorsing in hopes of causing economic or reputational consequences (poor review, low rating, uninstalls) to the company, forcing them to disassociate with the said celebrity. This would potentially scare public figures from taking stances on controversial issues which challenges their narrative. We pick the case of an online boycott of an e-commerce application Snapdeal, initiated by the Indian Right on Twitter for our study. This boycott campaign involved calls to collectively uninstall, give poor ratings, and leave bad app reviews on App store, over their brand ambassador Aamir Khan issuing a controversial statement on increasing intolerance in India. As this statement didn’t sit well with the Indian Right, they sought to leverage reputational harm as a way to force Aamir Khan to ”apologise” for it, and also to create precedence to scare public figures on challenging the right-wing narrative in the country. We conduct a mixed methods investigation to unpack co-ordinated user practice and behaviour in such online campaigns. We investigate how platform features get appropriated to serve the goals of the campaign, and how tweet framing strategies were effective in mobilising and calling to action. Through this thesis we make two contributions, 1) provide empirical results on user behaviour and tactics on mobilising for online collective action 2) theorise and present an argument how these campaigns are able to achieve very similar user goals as platforms designed for collective action do. We conclude with implications for design on building better online civic systems that rely on engaging large number of people and, how designers need to be cognisant of appropriation of features lest they become a moderation challenge.

Abstract

The ability to successfully imitate natural processes is vital for understanding the dynamic behaviour of atoms and molecules. These include reaction rates, mechanisms, energy configurations and macroscopic thermodynamic properties. For any such simulation, it is vital that a potential is constructed which can replicate the dynamics as closely as possible. Widely used classical potentials are constructed by fitting certain parameters with experimental results and then these potentials are used to run simulations under varied conditions. A drawback of this approach is its limited transferability, so the potential might not be suitable to simulate a system under conditions that differ vastly from the parameterization conditions, for example different solvents, temperatures, pressures etc. The accuracy of an empirical force field also depends on the quality of the parameterization that was done based on various approximations and the different sources of data used. The computationally expensive nature of ab initio molecular dynamics simulations severely limits its ability to simulate large system sizes and long time scales, both of which are necessary to imitate experimental conditions. In this work, we explore an approach to make use of the data obtained using the quantum mechanical density functional theory (DFT) on small systems and use deep learning to subsequently simulate large systems by taking liquid argon as a test case. A suitable vector representation was chosen to represent the surrounding environment of each Ar atom, and a ∆-NetFF machine learning model where, the neural network was trained to predict the difference in resultant forces obtained by DFT and classical force fields was introduced. Molecular dynamics simulations were then performed using forces from the neural network for various system sizes and time scales depending on the properties we calculated. A comparison of properties obtained from the classical force field and the neural network model was presented alongside available experimental data to validate the proposed method.

Abstract

Reading and writing documents is one among the primary skills with which we gather and communicate information. With the emergence of artificial intelligence (AI), researchers are in constant pursuit to build intelligent algorithms that can bring our physical and digital worlds close to each other. One such important domain is document image analysis, where we delve into the problem of understanding content from scanned document image collections. Considering “words” as the basic unit in understanding a document, in this thesis, we address the problem of finding the best possible representation for word images. Representation learning has been a key investigation for an AI problem. The primary goal of this thesis is to learn efficient representations for word images that encode its content. An ideal representation should be invariant to multiple fonts, handwritten styles and less sensitive to noise and distortions. In the past, representations have been handcrafted, specific to modalities (printed, handwritten), and sensitive to the complexities in handwriting in multi-writer scenarios. In this work, we choose the paradigm of learning from data using deep neural networks. We take our inspiration from the fact that given large amounts of annotated data, modern deep neural networks can inherently learn better representations. In this thesis, we also relax the need for large annotated datasets by heavily capitalizing on synthetically generated images. We also introduce a novel problem of learning semantic representation for word images which encodes the semantics of the word and reduces the vocabulary gap that exists between the query and the retrieved results. The first contribution of this thesis is a simple technique to generate large amounts of synthetic data, useful for pre-training deep neural networks. This led to the creation of IIIT-HWS dataset which is now widely used in the document community. The other major contributions of this thesis are: (a) the design of a deep convolutional architecture (named as HWNet) for learning an efficient holistic representation for word images, (b) a joint embedding scheme to project words and textual strings onto a common subspace, and (c) a novel form of word image representation which respects the word form along with its semantic meaning. The learned representations are evaluated under the tasks of word spotting and word recognition. We report state-of-the-art performance on popular datasets under both modern/historical and handwritten/printed document images while keeping the representation size compact in nature. Finally, in order to validate the proposed representations of this thesis, we present some interesting use cases such as (i) finding similarity between a pair of handwritten documents images, (ii) searching for keywords from online lecture videos, and (iii) building word retrieval system for Indic scripts.

Abstract

Neural Machine Translation (NMT) is a rapidly advancing MT paradigm and has shown promising results for many language pairs, especially in large training data scenarios. But Indian to English language Machine Translation is a challenging problem, owing to multiple factors including the structural and morphological difference, in addition to the lack of sufficient parallel training data, thereby demanding efficient strategies to improve the translation quality of the NMT systems. Although NMT is a promising approach, it still lacks the ability of modeling deeper semantic and syntactic aspects of the language. In machine translation with a low-resource setting, resolving data sparseness and semantic ambiguity problems can help improve its performance. In this thesis, we investigate utilizing extra syntactic and semantic linguistic factors in the context of the NMT framework for a low resource language pair i.e. Hindi-English. We propose a new architecture to incorporate explicit linguistic input features into the state-of-the-art Transformer network and demonstrate considerable performance improvements. Despite the massive success brought by neural machine translation, it has been noticed that the vanilla NMT often lags behind conventional machine translation systems, such as statistical phrase-based translation systems, for low-resource language pairs. In the past few years, various approaches have been proposed to address this issue but not much work has been done for Indian languages in this context. In this thesis, we also present our efforts towards building efficient Neural Machine Translation systems between Indian languages (specifically Indo-Aryan languages) and English via exploring the effectiveness of Multilingual Learning and Transfer Learning. We describe techniques to leverage the language relatedness among Indo-Aryan languages to improve the translation quality for individual language pairs. We also present our new approach Multilingual Transfer Learning which outperforms the aforementioned techniques. Neural MT models are generally trained using a Maximum Likelihood Estimation (MLE) objective and are tested with sequence level evaluation metrics such as BLEU. To address this inconsistency issue, Reinforcement Learning (RL) methods have been adopted to directly optimize sequence-level objectives. In this thesis, we also present an approach for training Neural Machine Translation systems using Advantage Actor-Critic method from Reinforcement Learning. Our approach directly optimizes the model parameters with respect to the task specific scores, unlike conventional maximum likelihood estimation and is fit for problems with low resource settings, large action space & delayed rewards. We also demonstrate experiments to leverage our approach to further boost the performance of NMT systems using source & target monolingual data for a low resource language pair like Hindi-English.