Digital Commons Network^™

Towards End-To-End Semi-Supervised Deep Learning For Drug Discovery, Xiaoyu Zhang

Computer Science and Engineering Theses

Observing the recent progress in Deep Learning, the employment of AI is surging to accelerate drug discovery and cut R&D costs in the last few years. However, the success of deep learning is attributed to large-scale clean high-quality labeled data, which is generally unavailable in drug discovery practices. In this thesis, we address this issue by proposing an end-to-end deep learning framework in a semi supervised learning fashion. That is said, the proposed deep learning approach can utilize both labeled and unlabeled data. While labeled data is of very limited availability, the amount of available unlabeled data is generally huge. …

Topological And Feature Based Identification Of Hole Boundaries In Point Cloud Data And Differentiation Between Surface And Physical Holes, Aaqif Muhtasim

Computer Science and Engineering Theses

With the advent of autonomous agents becoming prominent in everyday lives, the importance of processing the surroundings into understandable features becomes more and more important. 3D point clouds play a major role in the perception of such agents and thus having the ability to correctly decipher features from point clouds is crucial to the planning of actions that the agent would need to undertake. This thesis analyzes holes found in point clouds. Based on two approaches that center around topological data analysis and local point set features respectively. It studies how each of the methods works and how a combination …

Monitoring Of Swt2 Data Clusters For The Atlas Experiment, Antara Ray

Computer Science and Engineering Theses

Monitoring of the South West Tier 2 RSEs is done by CERN with the help of Rucio. The challenge faced by the team monitoring the servers at the University of Texas site was that the monitoring data is pictorially represented and provided to them in GIF format. In this work we focus on creating an interactive site that will not only monitor the data at the local RSEs but also create a platform to analyze the data storage systems. It turn it will also create alerts whenever during monitoring an aberration from expected behavior is noticed either in the storage …

Generating An Adaptive Path Using Rrt Sampling And Potential Functions With Directional Nearest Neighbors, Sandeep Chahal

Computer Science and Engineering Theses

Planning algorithms have attained omnipresent successes in several fields including robotics, animation, manufacturing, drug design, computational biology and aerospace applications. Path Planning is an essential component for autonomous robots. The problem involves searching the configuration space and constructing a desired collision-free path that connects two states (the start and the goal) for a robot to gradually navigate from one state to another. In global path planners, the complete path is computed prior to the robot set off. Sampling based planning like Rapidly Expanding Random Trees (RRT) and Probabilistic Road Maps (PRM) used for single or multi-query planning has gained popularity …

Large-Scale Deep Learning With Application In Medical Imaging And Bio-Informatics, Zheng Xu

Computer Science and Engineering Dissertations

With the recent advancement of the deep learning technology in the artificial intelligence area, nowadays people's lives have been drastically changed. However, the success of deep learning technology mostly relies on large-scale high-quality data-sets. The complexity of deeper model and larger scale datasets have brought us significant challenges. Inspired by this trend, in this dissertation, we focus on developing efficient and effective large-scale deep learning techniques in solving real-world problems, like cell detection in hyper-resolution medical image or drug screening from millions of compound candidates. With respect to the hyper-resolution medical imaging cell detection problem, the challenges are mainly the …

Building A Versatile Deduplication System, Zhichao Yan

Computer Science and Engineering Dissertations

With the development of the Internet and information technology, a large amount of unstructured data is generated and stored in various storage systems. In particular, data reduction techniques such as compression and deduplication have become an effective way to address the combined challenges of explosive growth in data volume but lagging network bandwidth growth to increase the space and bandwidth efficiency of various storage systems. However, we have found that existing deduplication systems cannot effectively process compressed data and image data because existing deduplication systems only analyze the hash value of the bitstream to detect redundant data. At the same …

Health Monitoring Of Atlas Data Center Clusters And Failure Analysis, Meenakshi Balasubramanian

Computer Science and Engineering Theses

Monitoring the health of data center clusters is an integral part of any industrial facility. ATLAS is one of the High Energy Physics experiments at the Large Hadron Collider (LHC) at CERN. ATLAS DDM (Distributed Data Management) is a system that manages data transfer, staging, deletions and experimental data on the LHC grid. Currently, the DDM system relies on Rucio software, with Cloud based object storage and No-SQL solutions. It is a cumbersome process in the current system, to fetch and analyze the transfer, staging and deletion metrics of a specific site for any regional center. In this thesis, a …

Defending Neural Networks Against Adversarial Examples, Armon Barton

Computer Science and Engineering Dissertations

Deep learning is becoming a technology central to the safety of cars, the security of networks, and the correct functioning of many other types of systems. Unfortunately, attackers can create adversarial examples, small perturbations to inputs that trick deep neural networks into making a misclassification. Researchers have explored various defenses against this attack, but many of them have been broken. The most robust approaches are Adversarial Training and its extension, Adversarial Logit Pairing, but Adversarial Training requires generating and training on adversarial examples from any possible attack. This is not only expensive, but it is inherently vulnerable to novel attack …

Text Mining On Twitter Data To Evaluate Sentiment, Srijanee Niyogi

Computer Science and Engineering Theses

Social media platforms have been a major part of our daily lives. But with the freedom of expression there is no way one can check whether the posts/tweets/expressions are classified on which polarity. Since Twitter is one of the biggest social platforms for microblogging, hence the experiment was done on this platform. There are several topics that are popular over the internet like sports, politics, finance, technology are chosen as the source of the experiment. These tweets were collected over a span of time for more than 2 months via a cron job. Every tweet can be divided into three …

Deepsign: A Deep-Learning Architecture For Sign Language, Jai Amrish Shah

Computer Science and Engineering Theses

Sign languages are used by deaf people for communication. In sign languages, humans use hand gestures, body, facial expressions and movements to convey meaning. Humans can easily learn and understand sign languages, but automatic sign language recognition for machines is a challenging task. Using recent advances in the field of deep-learning, we introduce a fully automated deep-learning architecture for isolated sign language recognition. Our architecture tries to address three problems: 1) Satisfactory accuracy with limited data samples 2) Reducing chances of over-fitting when the data is limited 3) Automating recognition of isolated signs. Our architecture uses deep convolutional encoder-decoder architecture …

Dwrelu : Double Weighted Rectifier Linear Unit An Activation Function With Trainable Scaling Parameter, Bhaskar Chandra Trivedi

Computer Science and Engineering Theses

Deep Neural Network have become very popular for computer vision application in recent years. At the same time, it remains important to understand the different implementation choices that need to be made when designing a neural network and to thoroughly investigate existing and novel alternatives for those choices. One of those choices is the activation function. The ReLU activation function is a widely used activation function. It discards all the values below zero and keeps the ones greater than zero. Variations such as Leaky ReLU and Parametric ReLU do not discard values, so that gradiants are nonzero for the entire …

Classification Of Clinical Narratives Using Convolutional Neural Network, Nikit Rajiv Lonari

Computer Science and Engineering Theses

Patient safety is a key aspect for good consumer care. When an individual is hospitalized or receives medication the family wants the patient safety to be above all factors. For instance, a drug can do both either cure the disease or perhaps, give rise to an adverse event. A drug administered for an indicated condition has substantial power to reduce or cure a disease and further to prevent it from happening again in the future but at the risk of side effects. At present, there are several methods in patient safety and in particular in the area of signal detection …

Krimet Kibernetike, Dua Gjyshinca

Theses and Dissertations

Në ditët e sotme moderne ku po mbizotëron e ashtëquajtura Epoka e Internet-it shumë nga punët e përditshme si pagesat, komunikimi në mes njerëzve, ndjekja e kurseve mësimore, shkarkimi dhe ngarkimi i dokumenteve të ndryshme, njoftimet rreth ngjarjeve në botë apo edhe përdorimi i aplikacioneve të ndryshme për argëtim ndodhin online me anë të pajisjeve të ndryshme si kompjuterët personal, telefonat e mençur dhe pajisjet tjera.

Njëra ndër gjërat më të rëndësishme që është e mundur falë internetit është mundësia që çdo kush në botë pavarësisht vendndodhjes i jepet mundësia që të lexojë dhe të njoftohet për ngjarjet nga e …

Algorithms For Exploratory Queries Over Web Database, Md Farhadur Rahman

Computer Science and Engineering Dissertations

In recent years we have seen an increase in the popularity of many web applications. The functionality of these applications range from allowing users to interact using online social network, to assist users in their everyday activity such as selecting a hotel in an area, locating a nearby restaurant etc. Google Maps, WeChat, FourSquare, AirBnB, TripAdvisor, and Hotels.com are a few such examples. The backed database of these applications can be a rich source of information for the corresponding application domain. For example, using Google Maps a user can find the ratings, reviews, and price of a restaurant, using Zillow …

Finding Representative Entities From Entity Graph By Using Neighborhood Based Entity Similarity, Ankit Anil Shingavi

Computer Science and Engineering Theses

Several applications deploy the use of large entity graphs. Given the entirety of its application scope, it is challenging to select a single entity graph for a particular need from numerous data sources. For a comprehensible overview of the entity graph, we may project a preview table for compact representation of an entity graph. Each preview table represents a single entity type in the dataset. We need to find the representative entities for a given entity type from the entity graph to show the coverage of a dataset. In this paper, we propose a method to find representative entities for …

Artificial Intelligence For Cognitive Behavior Assessment In Children, Srujana Gattupalli

Computer Science and Engineering Dissertations

Cognitive impairments in early childhood can lead to poor academic performance and require proper remedial intervention at the appropriate time. ADHD a?ects about 6-7% of children and is a psychiatric neurodevelopmental disorder that is very hard to diagnose or tell apart from other disorders. Cognitive insu?ciencies hinder the development of working memory and can a?ect school success and even have long term e?ects that can result in low self-esteem and self-acceptance. The main aim of this research is to investigate development of an automated and non-intrusive system for assessing physical exercises related to the treatment and diagnosis of Attention De?cit …

Improving Performance And Security In Anonymity Systems, Mohsen Imani

Computer Science and Engineering Dissertations

Tor is an anonymity network that provides online privacy for the Internet users. Tor hides the user's traffic among the others' traffic. The more users Tor attracts, the stronger anonymity it provides. Unfortunately, users of the Tor anonymity system suffer from less-than-ideal performance, in part because circuit building and selection processes are not tuned for speed. Moreover, there are some attacks like guard fingerprinting and website fingerprinting attacks that try to profile or de-anonymize the Tor users. In this dissertation, we propose methods to address both security and performance issues in Tor. We first examine the process of selecting among …

Malware Early-Stage Detection Using Machine Learning On Hardware Performance Counters, Anchal Raheja

Computer Science and Engineering Theses

Systems affected by Malware in the past 10 years has risen from 29 million to 780 million, which tells us it’s a rapidly growing threat. Viruses, ransomware, worms, backdoors, botnets etc. all come under malware. Ransomware alone is predicted to cost $11.5 billion in 2019. As the downtime and financial damages are rising the researchers are finding new ways to tackle this threat. However, the usual approach is prone to high false positive rate or delayed detection rate. This research explores a dynamic approach for early-stage malware detection by modeling it’s behavior using hardware performance counters with low overhead. The …

Supporting Efficient Large-Scale Key-Value Systems With An Optimized Storage Hierarchy, Xingbo Wu

Computer Science and Engineering Dissertations

Driven by the growing demands from big-data applications, the focus of their data management has been largely shifted from traditional SQL databases to NoSQL (Not-only- SQL) databases, such as key-value (KV) stores, which provides essential functionalities and much higher performance for storing and retrieving data. Correspondingly new hardware technologies have been developed to support the fast data accesses, such as NVMe SSDs and Infiniband network. However, existing designs of NoSQL databases usually see sub- optimal performance on fast hardware. Traditionally the computing overhead of a database system is overshadowed by the slow storage and network. With the adoption of the …

Hyper-Optimized Machine Learning And Deep Learning Methods For Geo-Spatial And Temporal Function Estimation, Neelabh Pant

Computer Science and Engineering Dissertations

Owing to a high degree of freedom in human mobility, accurate modelling/estimation of human mobility function remains a challenge. Numerous work in the literature have tried to address the challenge using various traditional machine learning methods on spatio-temporal attributes of data. We compare the use of Varied-K Means clustering, Hidden Markov Model techniques, feed forward neural networks, recurrent neural networks (RNN) and Long Short Term Recurrent Neural Networks (LSTM) to predict a user's future movement based on the user's past historical data. Although several techniques were proposed to predict a user's movement, not many have concentrated on a user's location …

Improving Time And Space Efficiency Of Trie Data Structure, Nirmik Milind Kale

Computer Science and Engineering Theses

Trie or prefix tree is a data structure that has been used widely in some applications such as prefix-matching, auto-complete suggestions, and IP routing tables for a long time. What makes tries even more interesting is that its time complexity is dependent on the length of the keys inserted or searched in the trie, instead of on the total number of keys in the data structure. Tries are also strong contenders to consider against hash tables in various applications due to two reasons - their almost deterministic time complexity based on average key length, especially when using large number of …

Face Detection And Recognition Using Moving Window Accumulator With Various Deep Learning Architecture, Anil Kumar Nayak

Computer Science and Engineering Theses

Recent advancement in the field of Computer Vision and Deep Learning is making object detection and recognition easier. Hence, growing research activities in the field of deep learning are enabling researchers to find new ideas in the area of face detection and recognition. Implementation of such systems has a number of challenges when it comes to the current approaches. In this paper, we have presented a system of Face Detection and Recognition with newly designed deep learning classification models like CNN, Inception and various state of art models like SVM and we also compared the result with FaceNet. Multiple approaches …

Design Of Haptically Enabled Wheelchair For Assistive Autonomy, Arjun Mani Gupta

Computer Science and Engineering Theses

The first records of wheeled seats being used for transporting disabled people date to 8th century in China, however the wheelchair has evolved tremendously since its inception. An electric-powered wheelchair, commonly called a "powerchair" is a wheelchair which incorporates batteries and electric motors into the frame, and so it can be controlled by either the user or an attendant. This control is most commonly done via a small joystick mounted on the armrest, or on the upper rear of the frame. For users who cannot manage a manual joystick, head-switches, chin-operated joysticks, sip-and-puff controllers or other custom controls may allow …

Enabling Third Party Services Over Deep Web Databases And Location Based Services, Yeshwanth Durairaj Gunasekaran

Computer Science and Engineering Theses

Deep web databases are pillars of today’s internet services hidden behind HTML forms and Top-K search interfaces. While Top-K search interfaces provide a good way to retrieve information, it still lacks in addressing the diverse preferences of the users. Due to query rate limit constraint - i.e., maximum number of k-Nearest Neighbors queries a user/IP address can issue over a specific period of time, it is often impossible to access all the tuples in backed database. With the query rate limit constraint in mind, our motivation is twofold (i) Enable users to obtain individual records from these databases and rank …

Hierarchical Representation Learning With Connectionist Models, De Wang

Computer Science and Engineering Dissertations

To unleash the power of big data, efficient algorithms which are scalable to millions of data are desired. Deep learning is one area that benefits from big data enormously. Deep learning uses neural networks to mimic human brains, this approach is termed connectionist in AI community. In this dissertation, we propose several novel learning strategies to improve the performance of connectionist models. Evaluation of a large neural network during inference phase requires a lot of GPU memory and computation, which will degrade user experience due to response latency. Model distillation is one way to distill the knowledge contained in one …

Interactive Learning And Adaptation For Personalized Robot-Assisted Training, Konstantinos Tsiakas

Computer Science and Engineering Dissertations

Robot-Assisted Training (RAT) is a growing body of research in Human-Robot Interaction (HRI) that studies how robots can assist humans during a physical or cognitive training task. Robot-Assisted Training systems have a wide range of applications, varying from physical and/or social assistance in post-stroke rehabilitation to intervention and therapy for children with Autism Spectrum Disorders. The main goal of such systems is to provide a personalized and tailored session that matches user abilities and needs, by adjusting task-related parameters (e.g., task difficulty, robot behavior), in order to enhance the effects of the training session. Moreover, such systems need to adapt …

Learning To Generate Individual Data Sequence From Population Statistics Using Dynamic Bayesian Networks, Mohammed Azmat Qureshi

Computer Science and Engineering Theses

Data collection rose exponentially with the dawn of the 21st Century, However the most important data to humans, individual health data, is difficult to get approved for public research, as medical history is very sensitive to be distributed. The only available public data which can be retrieved from institutions like the Centre for Disease Control (CDC), World Health Organization (WHO), National Health Interview Survey (NHIS), etc. largely only contain population statistics for different attributes of a person.What we propose here is a generative model which would learn to create data sequences for a population, each sequence mimicking an individual person’s …

Mavvstream: Expressing And Processing Situations On Videos Using The Stream Processing Paradigm, Mayur Arora

Computer Science and Engineering Theses

Image and Video Analysis (IVA) has been ongoing for several decades and has come up with impressive techniques for object identification, re-identification, activity detection etc. A large number of techniques have been developed and used for processing video frames to detect objects and situations from videos. Camera angles, lighting effect, color differences, and attire make it difficult to analyze videos. Several approaches for searching, and querying videos and images have been developed using indexing and other techniques. This thesis takes a novel approach by converting a video (through extraction of its contents) into a representation over which queries can be …

Indexing, Querying, Prediction, And Integration For Network-Constrained Moving Objects Databases, Mohammadhani Fouladgar

Computer Science and Engineering Dissertations

The emergence and presence of satellites and GPS devices have led to the creation of a huge amounts of spatial and spatio-temporal data, which had significant effects on creating new applications to analyze and mine these data. In this regard, a lot of research has been done on moving objects databases as a part of spatial and spatio-temporal databases. In this dissertation, we focus on those moving objects that are not allowed to move in all directions freely, but they (almost) always are restricted to travel on a specific network. One of the most popular example of these moving objects …

Efficient Evaluation Of Contextual And Reverse Pareto-Optimality Queries, Afroza Sultana

Computer Science and Engineering Dissertations

Many real-world applications analyze data to find objects that ``stand out'' with regard to various contexts and ways of valuing the objects. Examples of such application scenarios include vendors recommending products to potential customers, social networks improving content selection for users, and Google Scholar notifying newly published articles based on profiles. Besides, journalists identify conditions to substantiate the significance of an event or the interestingness of an object. Interesting events can be retrieved from stock data, weather data, and criminal records. Apart from journalists, those events convey significant information for financial analysts, scientist, and citizens. The aforementioned application needs can …