Abstract

Mobile robots have the potential to be widely used. They can be used in complex indoor and outdoor environments accomplishing jobs which typically are earmarked for humans. A basic requirement for these kind of robots is ability to navigate autonomously. For autonomous navigation to be possible the robot needs to be able to perceive the environment around; specifically localize itself in it, and create a map which it can use to plan and navigate. Localization and mapping are inter-dependent on each other, that is to say localization quality determines mapping quality and mapping quality determines further localization quality. Furthermore, it will determine how reliably we avoid obstacles, search for objects, determine the best path for a goal etc. Quality of the map which is created is of enormous importance, as it inherently dictates merit of the task robot is expected to execute. Accurate mapping of environment becomes more challenging when it is attempted using a monocular camera. Monocular camera posses it’s host of difficulties, but is very cost effective compared to expensive laser scanners and stereo cameras. As a sensor, monocular camera provides plethora of information like depth, color, texture which can be used object identification, face recognition, lip reading, traffic signs, and and 3D structure, most of which is still not being harnessed, and hence presents itself as a sensor for the future and worth pursuing. Therefore, we have used a Monocular camera as the quintessential sensor for mapping the surroundings. We are a long way from realizing monocular camera’s potential. As of now it is incapable of handling violent motions. A fast moving robot or a sharp turn leads to breaking of the mapping process. Sideways motion for the camera is an ideal movement for VSLAM systems, but this luxury is not afforded to mobile robots. Looking straight ahead and moving in front (along the principle axis) is a feasible motion for them. For such motion, moving straight in a line at slow speed(not too slow) produces better result, and when robot needs to turn it is done so moderately. Hence, we have proposed a planning framework which devices a path ensuring motion with aforementioned constraints, that is we have optimized heading change, and velocity with respect to distance from the goal. We model our robot as a non-holonomic vehicle which allows us to define feasible motion states for the robot. We than parametrize it’s states, namely it’s position and orientation into polynomial functions, which serve as easy inputs for optimization methods. We use optimization at two stages, first a non-linear optimization to generate path for a given start and goal position, and second a relatively naive quadratic optimization method to estimate initial guess for the non-linear optimizer. A good initial guess from the quadratic optimization method will be reflected in quality of final output. We show that by following the path generated by aforementioned framework VSLAM is able to localize camera successfully, which is to say estimated camera trajectory has similar curves to the generated trajectory. We also show following the generated constrained trajectory tracking performance improves in comparison to when robot follows unconstrained trajectory generated by RRT, one of the most popular algorithms for path planning. We add dense reconstruction of planar objects after the path planning framework in our attempts towards automation. Algorithms used to create 3D maps from stream of images have primarily focused on sparse point maps. Sparse maps are incomplete representation of the environment, making comprehension of available data very hard. They are also not able to cater complete information to modules down the pipeline of a full fledged robotic system. We have proposed a framework for dense reconstruction of 3D planar objects. Most objects in an indoor environment are planar, hence we have focused our attention towards reconstructing such surfaces. Targeting planar objects allows us to perform dense tracking in real time on commodity laptop. We have used Homography and segmentation techniques to identify and track planar objects. Post-reconstruction we present a modified Bundle Adjustment, an optimization technique used to correct position of 3D points. Traditionally VSLAM based systems have required human intervention, dense model of an environment will provide us with information such as span of obstacles and ground required for autonomous navigation. In results section we qualitatively demonstrate superiority of dense models over sparse point clouds. We also, both quantitatively and qualitatively show performance enhancement achieved by proposed bundle adjustment over original bundle adjustment. In conclusion, we present a path planning framework which allows robot to move by on it’s own accord if answer to “Where to go?” is known. We also present a reconstruction framework which improves

Abstract

Several natural phenomena manifest themselves as spatio-temporally evolving processes, for example climate patterns, disease spread and urban sprawl. The study of these processes, which aims to increase our understanding of the spatio-temporal phenomena for their prediction and control, requires analysis tools to infer models and their parameters from collected data. Whilst several studies exist on how to model the characteristics of spatio-temporal processes from highly complex patterns, a generic, statistically valid approach which may be readily employed in a wide range of scenarios, such as with systems with different forms of observation processes or time-varying systems, is lacking. This work attempts to fill this void by providing a systems approach to spatiotemporal modeling which can be used with continuous observations, point process observations, systems exhibiting spatially varying dynamics and time-varying systems.
Even though many works have been successful in mining spatio-temporal patterns from a Data Mining (DM) and Knowledge Discovery (KD) approach, they suffer from numerous limitations - they are tightly dependent on the ability to hypothesize, to determine the relevant factors and to deduce a conclusion, which require large thematic data and understanding of the subject to even define the primary hypothesis of the analysis. Hence, are heavily coupled to the domain. Moreover, they tend to neglect the continuous nature of the phenomenon they are trying to observe by discretizing the space or the time in which the co-location and co-occurrence pattern mining is attempted. Lastly, scale factors bring in more challenges which limit the model design and applicability, for instance, the occurrence of natural break lines can sub-divide or alter the choice of the appropriate scale of analysis. Therefore, there is an increasing need of developing an approach that can produce valid analyses of variations in dynamic phenomena, with particular focus on utilization of the inherent spatio-temporal characteristics of the data to address the issues mentioned above.
The developed set of methods act as a framework that leverages the observed event data - both spatial and temporal characterizations of observed event occurrences, making use of inherent spatio-temporal dependency in such data. A risk map is generated that provides valuable insights into its geographical spread and to help quantify the spatial risk factor associated with it, acting as a basis for dynamic spatiotemporal modeling which can easily cater for spatially varying dynamics. The work finally leads to the development of a novel method to model meso-scale processes by capturing spatio-temporal correlation leveraging both the potential dependencies among the local coefficients, and the potential incongruities within global models for prediction of trends. In the context of disease management, the proposed method is able to infer causality, and simulate the spread given the initial sources derived primarily from the spatio-temporal history of the disease.
The methods are seen to compare favorably with other conventional approaches and to work well on real temporal data sets. The potential of adopting the proposed approach to spatio-temporal modeling is seen in the illustrative case studies of Salmonellosis disease in three states of USA, which have been presented to demonstrate the utility of the method, where it is found that reliable predictions are possible even in complex scenarios. The encouraging results mimicing the observed trends are a strong indication that the adopted approach, per se may be used for large-scale spatio-temporal systems across several disciplines that exhibit similar spatio-temporal dynamic behavior and thus, provide a mechanism by which stochastic models are made available for spatio-temporal decision-making processes. It is evident that such a framework will help prioritize decision-making process for better risk assessment and management including disease outbreak, as shown in the illustrative case studies.

Abstract

Sentiment Analysis is an area of focus over the last decade. Increase in user-generated content provide an important aspect for the researchers, industries and government(s) to mine this information. The user-generated content is one important source for various organizations to know/learn/identify the general expression/sentiment of different users on the product. In this work, we focus on mining sentiments and analyzing them for Hindi language. Hindi is the 4th commonly spoken language1 in the world. With the increase in the amount of information being communicated via regional languages like Hindi, comes a promising opportunity of mining this information. Mining sentiments in Hindi comes with their share of issues and challenges. Hindi is morphologically rich and is a free order language as compared to English, which adds complexity while handling the user-generated content. The scarcity of resources for the Hindi language brings challenges ranging from collection and generation of datasets. We take up this challenge and work towards building resources- reviews, blogs annotated corpora and subjective lexicon for Hindi language. We propose a technique to build a subjective lexicon given a pre-annotated seed list for a language and its wordnet representing the network/connectivity of words using synonyms and antonyms relations. One of the salient features of this technique is that the method can be applied for any language which has the wordnet available. To show the applicability of the technique on other languages, we experiment our technique on English language in addition to the Hindi language. The lexicon generated by our algorithm is evaluated using the following different metrics. 1. Comparing against Manual Annotation 2. Comparing against similar Existing Resources 3. Classification Performance In addition to resource creation, we take up the task of sentiment classification in Hindi Language. We work on two different genres of Hindi User- Generated Web Content- 1. Reviews 2. Blogs For both of these genres we present three different approaches for performing sentiment classification such as 1. Using Subjective Lexicon 2. N-Gram Method 3. Weighed N-Gram We aim at analysing the merits and demerits of each of the above approaches across the different genres for the sentiment classification task. We discuss in detail the problems and the issues while working with the user-generated content (reviews and blogs) in Hindi language. This research work, throws some light on the main differences between the User-Generated Content in English and Hindi language at linguistic and its representation level and the approaches followed to address the same. English language provides the option of leveraging the abundant resources and tools that have been developed in the past, the work for Indian Languages has just began since last decade and it is in early stage of research and development, so our focus has been on- “To effectively mine the subjective information from the user-generated content in Indian languages, overcoming the data scarcity challenges associated with such problems.”.

Abstract

Stroke is a chronic disease which often leads to death. Different medical imaging modalities enable diagnosis for stroke after the onset of symptoms. Time is of the essence during stroke analysis since the window of therapy is very small (< 3 hrs after the onset of symptoms). Recent clinical studies have shown the usefulness and significance of diagnosing stroke on the Diffusion Weighted Magnetic Resonance Imaging (DWI) scans of the brain in the early stages. Visual inspection of the DWI scans is difficult since multiple scans are acquired for a patient with varied contrast and the scans depict complementary information about the diffusion process in the brain. To make matters worse, the DWI scans are acquired at a very low resolution with poor signal to noise ratio (SNR) since the time of acquisition is significantly less (< 1 min) and are confounded by artifacts that mimic stroke lesions. Thus, an automated framework which can accurately capture the stroke lesions in the DWI data would assist the clinicians in a better diagnosis. This is focus of the thesis. Varying the acquisition parameter (b-value) generates different DWI scans with varied contrast. DWI with higher b-values provide improved sensitivity, conspicuity of stroke lesions and reduced artifacts at the cost of lower SNR. Along with the DWI scans, the Apparent Diffusion Coefficients (ADC) maps are also derived which give a measure of the true diffusion process in the brain irrespective of the acquisition artifacts that resemble stroke. In this thesis, we argue that integrating information from multiple sources, namely, low and high b-value data along with the ADC maps, can aid better characterization of stroke lesions in the data. Accordingly, we propose a novel approach for detecting and segmenting stroke regions from DWI data.

Abstract

Monocular robotic system capable of autonomous navigation that includes collision avoidance, object tracking and interaction with objects is indeed exciting for it paves way for replacing expensive sensors on-board robotic systems [2], such as 3D Laser, high end Global Positioning Systems and Attitude Head Reference Systems. Integral components of such navigation system include, the ability to estimate the vehicle's pose in dynamic environments, ability of the system to accurately segment and reconstruct moving objects. In this thesis, we present novel methods for accurate motion segmentation and accurate reconstruction of moving objects. In case of motion segmentation, we propose an incremental motion segmentation algorithm. The motion segmentation is robust and is capable of segmenting dicult degenerate motions, where the moving objects is followed by a moving camera in the same direction. We use ecient geometric constraints that helps in segmenting these degenerate motions. In case of motion reconstruction, we use our multibody VSLAM framework. The multibody VSLAM framework provides for reconstruction of moving and stationary world. However, to provide a unied representation, it requires a solution to the relative scale problem. This relative scale problem occurs in a multibody setting when the reconstructed points of the stationary world and moving object(s) are to be represented in a unied frame of reference. This relative scale problem has proven to be one of most challenging open problem in robotics-vision community. Even recent approaches based on the multibody VSLAM framework have only managed to reduce the space of possible scales but are unable to provide for the absolute scale. We propose a number of solutions to the relative scale problem that provide for scale resolution. In particular, we advanced the state-of-the-art by resolving scale within two views (previous approaches required more than six views) with minimal restrictions on object and camera motion. Two more solutions, one requiring three camera views for any outdoor motion and other requiring four camera views for non-holonomic motion, form important contributions of this work. High delity reconstruction of complex moving trajectories at accurate relative scales are provided on real world dataset and results on synthetic datasets establish the ecacy of the formulation. Performance gain with previous literature on the basis of reconstruction accuracy, prevalent degeneracies and the immediateness of the solution are highlighted in detail. These results advance the state of the art of monocular VSLAM systems in dynamic environments where the camera trajectory, stationary and moving objects are all estimated and further represented at the correct scales.

Abstract

In the present work, ab initio and density functional theory (DFT) calculations are employed to compare the ability of various hybrid GGA density functionals and dispersion corrected functionals in exploring structural, reactivity and electronic properties of linear and cyclic peptides containing tetrahydrofuran (THF) and furan amino acids. A theoretical study of stereochemical control in conformational preferences of the peptides is an important step in designing architecturally ordered structures of furanoid sugar amino acid based molecular frameworks. Also, using the functionals, conformational behavior of 2;3 and 3 monomers and their homodimers is investigated. The monomers are designed to gain control over cis/trans isomerism of proline in order to generate ordered polyproline based foldamers. Using B3LYP/6-31G(d,p) level of theory, we have studied the properties of terminally blocked linear tripeptide containing tetrahydrofuran amino acids (TAA), Boc-TAA-Leu-Val-OMe (TLV). Conformational space of the peptides with four possible combinations of stereochemistry at 2, 5 positions of the TAA, i. e., (2R,5S)-cis-TAA (TLV1), (2S,5R)-cis-TAA (TLV2), (2R,5R)-trans-TAA (TLV3) and (2S,5S)-trans-TAA (TLV4) is investigated. Stable conformer of TLV1 shows a three centered H-bond interaction. This is later conrmed by NMR and X-ray crystallographic studies. Conformations of other peptides are stabilized by regular H-bond interactions. The interactions are part of turn like structures. While a 10 membered -turn type structure is most likely found in TLV1-2, a 7 membered -turn type structure is predominant in TLV3-4. The structures stabilized in terms of thermodynamic and kinetic parameters are veri ed by experimental structures. The calculations on reaction of peptide bond formation between Boc-TAA and Leu-Val-OMe suggest that PBE0 functional reduces the electronic energy barrier heights by about 5 to 10 kcal/mol as compared to that measured by B3LYP functional. The stability of the peptides comes not only from intramolecular H-bonds but also from minimal entropic cost or antiparallel alignment of dipole moment vectors at adjacent amide bonds. Cyclization is a method to induce desired structural biases required for their biological properties. Thermodynamic and kinetic aspects of selective modes of cyclization of the linear tripeptides containing (2S,5R)-cis-TAA and (2S,5S)- trans-TAA is investigated using B3LYP/6-31G(d,p) level of theory. Electronic energy of activation for cyclization reactions were also studied using PBE0/6- 31G(d,p), M06-2X/6-31G(d,p) and MP2/6-31G(d,p) levels of theory. The calculations are also carried out at array of Pople type and Dunning basis sets with B3LYP functional. All the calculations show preference for intermolecular cyclization over intramolecular cyclization of the tripeptides. Amongst the three functionals, M06-2X shows low barrier heights. Greater barrier heights for the intramolecular cyclization are due to deviation in planarity of peptide bond. It makes the bond length longer which leads to less delocalization of lone pair electrons of oxygen with that on nitrogen. Stability for cyclic hexapeptides containing (2S,5R)-cis-TAA comes from a pair of thirteen, ten and seven membered H-bonded interactions. The other peptide having (2S,5S)-trans-TAA is associated with a pair of thirteen and ten membered interactions. Also, fractional increase in chemical hardness and decrease in polarizability in intermolecular cyclization is relatively larger compared to that in intramolecular cyclization. Our theoretical insights are supported by experimental results. Using B3LYP/6-31G(d,p) and M06/6-31G(d,p) levels of theory, we have investigated homo versus heterochiral cyclic trimerization of 5-(aminoethyl)-2- furancarboxylic acid in gas phase, implicit and explicit solvent phases. The cyclic peptides are useful tools for studying molecular recognition processes and show anion binding properties. Gas phase and PCM calculations show that homochiral cyclic peptide is thermodynamically favorable. We have envisaged that change in stereochemistry at one chiral center in the cyclic tripeptide changes symmetry in the structure, nature of intermolecular forces and its reactivity. The B3LYP calculations on solvent assisted structures show only thermodynamic favorability for heterochiral cyclic peptide. On the other hand, M06 calculations on the systems reveal both thermodynamic and kinetic preferences for the heterochiral cyclic tripeptide, compared to its homochiral counterpart. Change of stereochemistry at one chiral center makes the cyclic peptide amenable to interact with solvent molecules strongly from both of its faces. Further support comes from studies based on IR spectra, NBO, AIM and MESP plots. The study on another class of synthetic amino acids, i. e, 2;3- (monomer I), and 3- (monomer II) substituted homoproline reveals t

Abstract

The role of Internet in personal, economic and political advancement is growing in a fast pace. By the turn of century, data on web reaches to petabytes or exabytes or may even scale up-to unimaginable quantities. Extraction of precise and structured information from such large amounts of unstructured or semi-structured data is the major concern of web known as Information Extraction. Named entity recognition (NER) (also known as entity identification and entity extraction) is one of the important subtask of information extraction that seeks to locate and classify atomic elements in text into predefined categories such as the names of persons, organizations, locations, monetary values, percentages, expressions of times, etc. NER has many applications in NLP, for e.g., in data classification, question answering, cross language information access, machine translation system, query processing, etc. Recognizing Named Entities (NEs) in English has reached accuracies nearing to 98%. For English, many cues aid to know the structure of language (one such important cue in identifying NEs is capitalization) which made the accuracies to be high. Whereas in Indian languages, there are no such cues available and moreover each Indian language differ from the other in grammatical structure. Hence, developing a language independent NER is a challenging task. Previous works includes developing an NER system using language dependent tools such as POS Tagger, dictionaries, Chunk Tagger, gazetteer lists, etc., or they have used linguistic experts to manually tag the training and testing data or linguistic experts used to generate rules for recognizing NEs. Language Independent approaches include supervised machine learning techniques such as CRF, HMM, MEMM, SVM, etc. These techniques need High amounts of manually tagged data which is again a point of concern. Some of the other approaches include exploiting the external knowledge such asWikipedia. But, in those methods the utilization of Wikipedia is not complete. Hence, the main objective of this work is to build a language independent NER system without any manual intervention and without any usage of language dependent tools. The approach specified throughout the work, includes language independent methods to identify, extract and recognize the NEs. Identification of NEs is done using an External Knowledge namely Wikipedia. More specifically, English Wikipedia is used as an aid to derive the NEs from Indian languages. Wikipedia hierarchal structure is explored and the documents in it are divided into specific domains. Each domain is considered and the corresponding English and Indian language documents are clustered. English documents are tagged using the Stanford NER Tagger and the non-NEs are removed. Using the term co-occurrences between the tagged English and non-tagged Indian language words, the corresponding NEs between Indian language and English are mapped. Thus the tag of English NE is duplicated to the Indian language NE. Hence, the Indian language data is tagged. The tagged data generated in previous step, is used in recognition of NEs on sets of monolingual Indian language documents. In this step, a set of features are generated from the words of these documents and these features are used for recognition of NEs in a new document. Consider each document; extract the tagged data from the document using the data from previous step. Now, from the remaining words of the document, a Naive Bayes Classifier is build which uses these words to generate a set of features for each class (features here are nothing but the important words of a particular class in that document). The importance of these features is calculated statistically by different metrics (the metrics for classification). Now given a new document, the presence of these features along with their scores is calculated. If the score exceed a threshold, implies the presence of NEs in the document. By decreasing the size of document the process is repeated again till we get the NE. Hence, the monolingual Indian language document is tagged. The approach specified in identifying and recognizing the NEs is language independent and can be extended to any language as none of the language dependent tools are used or there is no involvement of linguistic experts. Hindi, Marathi and Telugu were the languages in which the work has been done. PERSON, LOCATION and ORGANIZATION were the tag of NEs used throughout the identification and recognition process. Wikipedia is used as a dataset in identifying the NEs. Around 3,05,574 English documents, Hindi 100,000 documents, Marathi 83,000 documents, Telugu 85,000 documents are used to generate the results. The results are evaluated on manually tagged 2328, 1658, 2200 Hindi, Marathi and Telugu Wikipedia documents respectively. The F-Measure scores are 80.42 for Hindi, 81.25 for Marathi and 79.98 for Telugu. Dataset for recognition of NEs is a

Abstract

For long it has been argued that metaphor is not merely a linguistic device used figuratively in poems and literature, but a conceptual process that plays a central role in cognition. Several attempts have been made to consider metaphor as a conceptual phenomenon (Ortony, 1979; Lakoff & Johnson, 1980). Most notably, Richards (1936) argued “that thought is metaphoric and proceeds by comparison, and the metaphor of language derive there from” (1936:94). Lakoff and his colleagues proposed conceptual theory of metaphor and argued that human cognition is organized in conceptual schemas, which are metaphoric in nature. These schemas are constantly enriched and modified by the interaction with the world. Language draws on these cognitive schemas but is not identical with them. Moreover, verbal metaphors are surface manifestations of the metaphorical grounding of these schemas. Therefore, metaphor is ‘fundamentally conceptual, not linguistic in nature’ (Lakoff, 1993). One of the crucial outcomes of this cognitivist paradigm is its implication that metaphor can assume non-verbal and multimodal manifestations as well. The focus of this thesis is on visual metaphors, which are pictorial counterparts of verbal metaphors such as “The sky is crying,” or “The man is a wolf.” Visual or pictorial metaphor is nonverbal manifestation of metaphorical thought, where one or both concepts of the metaphor (the target and the source) are depicted in images. In this thesis, our aim is to study different kinds of visual metaphors: How are they generated and how are they different from verbal metaphors. We focus on the difference between them at the presentation level, which relates to the early stages of perceptual processing. A verbal metaphor is presented as text but a visual metaphor is presented in images. Several psychological studies (Paivio, 1989; Schnotz & Bannert 1999) have shown that text and images are processed in different ways. Recent brain imaging studies have also confirmed that the right-hemisphere of the brain is specialized to process visual information (Bryden, 1982; Ellis et. al., 1988; Jonides et. al., 1993; Kosslyn et. al., 1993; McCarthy et. al., 1994; Smith et. al., 1995), whereas the left-hemisphere mostly processes linguistic or alphabetic information (Beaumont, 1982; Binder et. al., 1995; Desmond et. al., 1995; Gabrieli et.al., 1996; Howard et. al., 1992). We assume that this difference in perceptual processing is also reflected in high-level semantic processing tasks such as metaphor.

Abstract

Controlling molecular processes using lasers has been a long cherished dream since their invention in the 1960s. Of the several control schemes subsequently developed, Quantum Optimal Control Theory (QOCT) has been one of the most successful for the theoretical design of laser fields for driving molecular systems to accomplish specified objectives. In this work, we attempt a reformulation of the quantum optimal control problem for controlling vibrational motion in a molecule using an external laser field and under the constraint of reasonable fluence and evolution governed by the Schr¨odinger equation in Hilbert space. We do this by transforming the Schr¨odinger equation into a set of Hamilton’s equations in new phase space coordinates. We study the characteristics of dynamics in the K¨ahler space thus generated. We examine the geometric interpretation of this dynamics and reformulate the quantum optimal control problem as a classical control problem. We apply Pontryagin’s maximum principle to this classical control problem and explore its consequences for the quantum description. We examine in particular, controllability of the system and investigate how it changes in going from the original quantum problem to the classical-like reformulation. We speculate the role of adjoint variables in the reformulation as possible Lyapunov exponents in the classical-like phase space.

Abstract

A K -Nearest Neighbor based classifier classifies a query instance based on the class labels of its neighbor instances. Although kNN has proved to be a ubiquitous classification/regression tool with good scalability, but it suffers from some drawbacks. Two of its major drawbacks are: (1) The existing kNN algorithm is equivalent to using only local prior probabilities to predict instance labels, and hence it does not take into account the class distribution around the wider neighborhood of the query instance, which results into undesirable performance on imbalanced data. (2) It uses all the training data at the runtime and hence is slow. In this thesis, we provide a suite of solutions to tackle this drawbacks. BINER: We propose a new efficient technique for regression. Our algorithm instead of directly predicting the value of response variable, narrows down the range in which the response variable has the maximum likelihood of occurrence and then interpolates to give the output. The data is Hierarchically partitioned in the pre processing step, and search for the partition in which the response has the maximum likelihood of occurrence is carried out at the run time. It takes a single parameter k, the same as in kNN and more than often outperforms the conventional state of art methods on a wide variety of datasets as illustrated by our experimental study.