215 research outputs found
A higher order collective classifier:
Modern statistical machine learning techniques often rely on the assumption that data instances are independent and identically distributed (IID). However, recent work in statistical relational learning has demonstrated the utility of violating the independence assumption. Specifically, the research has shown the value of leveraging relationships between data instances based on higher-order paths. In this thesis, I present a novel Higher Order Collective Classifier (HOCC), a statistical relational machine learning technique that leverages latent information present in higher-order co-occurrences of items across data instances. A general framework is presented in which HOCC can be applied to event detection in time series data. Given the importance of cyber-security, HOCC is applied to two different data sets in the cyber-security domain: first, a Border Gateway Protocol (BGP) dataset, for detection and classification of anomalies, and second, a Network File System dataset for building models of user activity for masquerade detection. Performance of HOCC compares favorably against first-order models that do not leverage higher-order information, achieving separation of classes that heretofore were difficult to separate.M.S.Includes bibliographical references (p. 60-63)by Vikas Meno
Informetrics on M. N. Srinivas
M. N. Srinivas, the well known sociologist is widely recognised as architect of modern Indian sociology and social anthropology. His publications have been analysed by year, domain, authorship pattern, channels of communication used. Keywords, etc. The results indicate that the papers published by him are of a nature that qualify him to be a 'role model' for the younger generations to emulate.
By the end of 1995, Srinivas had to his credit 144 papers which, included 33 broad papers in sociology and anthropology; 18 papers in social change; 28 papers in village studies; 12 papers on religion; 17 papers on caste and 36 papers of general popular interest. The periods 1958-61 and 1974-77, when Srinivas was 38-41 and 58-61 years old. were his most productive periods with highest publication activity
Computational Complexity of Graph Partition under Vertex-Compaction to an Irreflexive Hexagon
In this paper, we solve a long-standing graph partition problem under vertex-compaction that has been of interest since about 1999. The graph partition problem that we consider in this paper is to decide whether or not it is possible to partition the vertices of a graph into six distinct non-empty sets A, B, C, D, E, and F, such that the vertices in each set are independent, i.e., there is no edge within any set, and an edge is possible but not necessary only between the pairs of sets A and B, B and C, C and D, D and E, E and F, and F and A, and there is no edge between any other pair of sets. We study the problem as the vertex-compaction problem for an irreflexive hexagon (6-cycle). Determining the computational complexity of this problem has been a long-standing problem of interest since about 1999, especially after the results of open problems obtained by the author on a related compaction problem appeared in 1999. We show in this paper that the vertex-compaction problem for an irreflexive hexagon is NP-complete. Our proof can be extended for larger even irreflexive cycles, showing that the vertex-compaction problem for an irreflexive even k-cycle is NP-complete, for all even k \geq 6
Assessment of the Toxic Effects of Silver Oxide Nanoparticles (Ag2O NPs) on the Biochemical Profile of Lissachatina fulica (Mollusca: Achatinidae)
Nanotechnology offers a promising approach for pest management, and this study evaluated the effects of silver oxide nanoparticles (Ag₂O NPs) on the biochemical composition of the giant African snail Lissachatina fulica. Snails were exposed to two concentrations (5 and 10 ppm) for 48 and 96 hours, and changes in protein, lipid, and carbohydrate levels were analyzed in the intestine and foot muscles.
The results showed a significant, concentration- and time-dependent decline in all biochemical components in both tissues. At 10 ppm after 96 hours, protein, lipid, and carbohydrate contents were markedly reduced, with the intestine showing greater sensitivity than the foot muscle. These findings indicate that Ag₂O NPs induce metabolic stress, leading to depletion of energy reserves and disruption of biochemical balance, highlighting the potential physiological risks of prolonged nanoparticle exposure
Ever changing information scenario and emerging role of national and regional associations in Indian library development
Library associations have had a presence in India since early nineties currently experiencing a remarkable increase in numbers and facing significant challenges as they move into digital future. An attempt has been made to comprehend the status of library associations in India through SWOT analysis and in this way this paper intends to identify the strengths & potential weaknesses and draw attention on the immediate need of restructuring and merging these associations for new working model should function in close partnership with other groups; could be stronger financially and have more clout in pursuing and meeting the objectives and goals of its members. Author has tried to explore the extent to which library associations seek to protect and advance the interests of members and responding to the ongoing changes and preparing them for the future. At the conclusion recommendations are made for carefully planned public relations programs by leveraging convergence Technologies could guarantee long- term success of library associations
Algorithms for Data-Efficient Training of Deep Neural Networks
Deep Neural Networks ("deep learning") have become a ubiquitous choice of algorithms for Machine Learning applications. These systems often achieve human-level or even super-human level performances across a variety of tasks such as computer vision, natural language processing, speech recognition, reinforcement learning, generative modeling and healthcare. This success can be attributed to their ability to learn complex representations directly from the raw input data, completely eliminating the hand-crafted feature extraction from the pipeline. However, there still exists a caveat: due to the extremely large number of trainable parameters in Deep Neural Networks, their generalization ability depends heavily on the availability of a large amount of labeled data.
In many machine learning applications, gathering a large amount of labeled data is not feasible due to privacy, cost, time or expertise constraints. Examples of such applications are abundant in healthcare; for example, predicting the effect of a medicine on a new patient in the scenario where the medicine has been administered to only a few patients earlier. This thesis addresses the problem of improving the generalization ability of Deep Neural Networks using a limited amount of labeled data. More specifically, this thesis explores a class of methods that directly incorporates the inductive bias about how the Deep Neural Networks should "behave" in-between the training samples (both in the input space as well as the hidden space) into the learning algorithms. Throughout several publications included in this thesis, the author has demonstrated that such kinds of methods can outperform conventional baseline methods and achieve state-of-the-art performance across supervised, unsupervised, semi-supervised, adversarial training and graph-based learning settings.
In addition to these algorithms, the author proposes a mutual information based method for learning the representations for the "graph-level" tasks in an unsupervised and semi-supervised manner. Finally, the author proposes a method to improve the generalization of ResNets based on the iterative inference view
Design of a working model of an upper limb prosthesis: wrist mechanism
This thesis demonstrates a new design for an upper limb prosthetic wrist that gives 3 independent degrees of freedom (DOFs) through individual mechanisms. A human wrist has 3 degrees of freedom i.e. Flexion-Extension, Radial- Ulnar deviation and Pronation-Supination. The upper limb prostheses that are currently available in the market generally provide 1 (usually Pronation- Supination) or at most 2 degrees of freedom, which is not su cient for daily life. For this thesis, a new wrist having all the 3 DOFs was designed in the SolidWorks software, a prototype was 3D printed and a basic analysis of the mechanical properties of the model through SolidWorks simulation was carried out. The prototype mechanisms were then connected to servo motors, with potentiometers as their inputs, that were programmed through an arduino and were tested to see if they work as expected. Faithful recreation of the wrist motions was achieved and the range of motion (ROM) of this prosthesis was similar to the ROM of an actual human wrist. This thesis also looks at the other prostheses that are available in the market or are under development and their limitations are discussed. The goal of this thesis is to present a design of an upper limb prosthesis which addresses some of the limitations while also encouraging further research and development in this eld and will hopefully result in more people with amputations choosing to use prosthetics in their daily lives.M.S.Includes bibliographical referencesby Sahil Vikas Dang
The Poverty in Q and A by Vikas Swarup: The Portrait of Exploitation of Indian underclass
ABSTRACT
Poverty is the one of the most crucial problems in the world history till today; it reveals in any forms of human’s deprivation suck lack of food, house, dress etc. It has made poor people in the suffered condition. Furthermore, the poor people in India live in the structural poverty that they will never change the better life because any community that can determine someone can be rich or poor. Thus phenomenon is reflected in the novel Q and A by Vikas Swarup as Indian. It tells the worst condition happened the poor condition in India where the poor get poorer and poorer on the contrary the rich is richer and richer
Thus, this study is aimed to know how the structural poverty that is described in the novel, what the causes and effects of it and the last is how the Q and A reflect the real condition in India. Then, to achieve the problem of the study, the researcher uses sociological literary criticism as one method of this research. Moreover, the data of this research are collected from Q and A by Vikas Swarup. The researcher herself becomes the research instrument in the process of data collection and data analysis.
Based on the problem of the study, the result finding and analysis of this research is divided into three main parts generally. The first is describing structural poverty which appeared as social problem in India. Then, then second is the causes and effects of structural poverty, the researcher found that most of the reasons of appearance of the poverty. The data tells that the reasons of the structural poverty is caused by the passiveness of the government in seeing the poor and the unfair market that is always dominated by the rich because they have much capital. Here, many portraits of unfair treatments in the real condition of the poor in India appears in the novel; the poor in India go to worst which it can be seen that many slums house in middle of the city with no facilities to support their life. The trafficking does not only happen to the poor children but also the poor girl in India. The unfair treatment of the rich is they are ordered to work but the money must be paid to the employer, the rich only use the poor as money engine that only give profit to the rich. Many poor children can not get school because of poverty.
Finally, The story of the structural poverty in India is portrait from independent of India until now, the rich dominate in many sectors and the government does not pay attention to the poor. That phenomenon was percieved by the author. It indicates how the background of the author affects the work itself. It means that the story of structural poverty in India was told based on the real social condition
Para além do exótico ou bélico: cinemas e histórias do país de Tagore e Jamal Malik
The author proposes reflections on the motion picture Who Wants to Be a Millionaire?, by Danny Boyle and screenplay by Simon Beaufoy, adapted from the novel by Vikas Swarup, Q & A. The objective is to recover aspects from the Hindu history and culture using the history of its cinemas.a autora propõe para esta edição da revista reflexões a partir do filme Quem quer ser um milionário?, de Danny Boyle, com roteiro de Simon Beaufoy, adaptado da novela de Vikas Swarup, Q & A. O objetivo é recuperar aspectos da história e da cultura indiana a partir dos seus cinemas
Characterization and optimization of photonic devices fabricated using femtosecond laser micro-matching
Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2005.Includes bibliographical references (leaves 84-90).There is a great deal of interest and activity in the area of femtosecond micromachining of transparent materials. It promises to be a powerful technique for rapid fabrication of photonic devices in three dimensional geometries. Our group has fabricated and investigated an array of two dimensional and novel three dimensional photonic devices that are able to perform their intended functions but whose loss properties are not yet well known. The first aim of this Masters thesis is to study waveguide loss. It will focus on studying and characterizing the losses of straight and circularly curved waveguides - the building blocks of many practical devices. With a proper understanding of the loss per unit length, per unit bend angle, and per re-write of the waveguides a structured set of guidelines for device fabrication for our particular setup can be made. The second aim of this thesis is the characterization of the photonic devices fabricated and demonstrated by the author. These include a broadband characterization of directional and X couplers, as well as a demonstration of 3D symmetrical 1 :N waveguide splitters for optical signal distribution. These experimental results and discussions, which form the core of the thesis, flow in a logical fashion - from the elemental straight waveguides to the curved waveguides that are used to design the directional couplers. The X couplers provide the broadband performance that the directional couplers cannot and the 1 :N waveguide splitters fabricated with the aid of an improved experimental setup not only provide a more elegant re-design of the broadband X couplers but also demonstrate scalability to N output ports in three dimensions.by Vikas Sharma.S.M
- …
