Physical Sciences and Mathematics Commons^™

Pattern Discovery In Dna Using Stochastic Automata, Shweta Shweta

Master's Projects

We consider the problem of identifying similarities between different species of DNA. To do this we infer a stochastic finite automata from a given training data and compare it with a test data. The training and test data consist of DNA sequence of different species. Our method first identifies sentences in DNA. To identify sentences we read DNA sequence one character at a time, 3 characters form a codon and codons form proteins (also known as amino acid chains).Each amino acid in proteins belongs to a group. In total we have 5 groups’ polar, non-polar, acidic, basic and stop codons. …

Logistic Regression Models To Predict Solvent Accessible Residues Using Sequence- And Homology-Based Qualitative And Quantitative Descriptors Applied To A Domain-Complete X-Ray Structure Learning Set, Reecha Nepal, Joanna Spencer, Guneet Bhogal, Amulya Nedunuri, Thomas Poelman, Thejas Kamath, Edwin Chung, Katherine Kantardjieff, Andrea Gottlieb, Brooke Lustig

Faculty Publications, Chemistry

A working example of relative solvent accessibility (RSA) prediction for proteins is presented. Novel logistic regression models with various qualitative descriptors that include amino acid type and quantitative descriptors that include 20- and six-term sequence entropy have been built and validated. A domain-complete learning set of over 1300 proteins is used to fit initial models with various sequence homology descriptors as well as query residue qualitative descriptors. Homology descriptors are derived from BLASTp sequence alignments, whereas the RSA values are determined directly from the crystal structure. The logistic regression models are fitted using dichotomous responses indicating buried or accessible solvent, …

Cooling Atomic Gases With Disorder, Ehsan Khatami, Thereza Paiva, Shuxiang Yang, Valéry Rousseau, Mark Jarrell, Juana Moreno, Randall Hulet, Richard Scalettar

Faculty Publications

Cold atomic gases have proven capable of emulating a number of fundamental condensed matter phenomena including Bose-Einstein condensation, the Mott transition, Fulde-Ferrell-Larkin-Ovchinnikov pairing, and the quantum Hall effect. Cooling to a low enough temperature to explore magnetism and exotic superconductivity in lattices of fermionic atoms remains a challenge. We propose a method to produce a low temperature gas by preparing it in a disordered potential and following a constant entropy trajectory to deliver the gas into a nondisordered state which exhibits these incompletely understood phases. We show, using quantum Monte Carlo simulations, that we can approach the Néel temperature of …

University Scholar Series: Michael Kaufman, Michael Kaufman

University Scholar Series

H2O in Interstellar Space: How the Universe Conspires to Make Water Everywhere

On October 28, 2015, Dr. Michael Kaufman spoke in the University Scholar Series hosted by Provost Andy Feinstein at the Dr. Martin Luther King, Jr. Library. His talk was titled “H₂O in Interstellar Space: How the Universe Conspires to Make Water, Water Everywhere.” Dr. Kaufman's astrophysics research focuses on the interactions and feedback between newly formed stars and the interstellar medium—the raw material from which stars form. He constructs computational models of the radiative transfer, dynamics and chemistry that occur in regions of active star formation, …

Topography And Tropical Cyclone Structure Influence On Eyewall Evolution In Typhoon Sinlaku (2008), Cheng-Hsiang Chih, Kun-Hsuan Chou, Sen Chiao

Faculty Publications, Meteorology and Climate Science

Typhoon Sinlaku (2008) was a tropical system that affected many countries in East Asia. Besides the loss of life and economic damage, many scientific questions are associated with this system that need to be addressed. A series of numerical simulations were conducted in this study using V3.2 of the advanced research version of the Weather Research and Forecasting (WRF-ARW) model to examine the impacts of different terrain conditions and vortex structures on the eyewall evolution when Sinlaku was crossing Taiwan. The sensitivity experiments using different vortex structures show that a storm of the same intensity with a larger eyewall radius …

Question Answering System For Yioop, Niravkumar Patel

Master's Projects

Yioop is an open source search engine developed and managed by Dr. Christopher Pollett. Currently, Yioop returns the search results of the query in the form of list of URLs, just like other search engines (Google, Bing, DuckDuckGo, etc.) This paper created a new module for Yioop. This new module, known as the Question-Answering (QA) System, takes the search queries in the form of natural language questions and returns results in the form of a short answer that is appropriate to the question asked. This feature is achieved by implementing various functionalities of Natural Language Processing (NLP). By using NLP, …

Graph Basesd Word Sense Disambiguation For Clinical Abbreviations Using Apache Spark, Veebha Padavkar

Master's Projects

Identification of the correct sense for an ambiguous word is one of the major challenges for language processing in all domains. Word Sense Disambiguation is the task of identifying the correct sense of an ambiguous word by referencing the surrounding context of the word. Similar to the narrative documents, clinical documents suffer from ambiguity issues that impact automatic extraction of correct sense from the document. In this project, we propose a graph-based solution based on an algorithm originally implemented by Osmar R. Zaine et al. for word sense disambiguation specifically focusing on clinical text. The algorithm makes use of proposed …

Clustering Web Concepts Using Algebraic Topology, Harleen Kaur Ahuja

Master's Projects

In this world of Internet, there is a rapid amount of growth in data both in terms of size and dimension. It consists of web pages that represents human thoughts. These thoughts involves concepts and associations which we can capture. Using mathematics, we can perform meaningful clustering of these pages. This project aims at providing a new problem solving paradigm known as algebraic topology in data science. Professor Vasant Dhar, Editor-In-Chief of Big Data (Professor at NYU) define data science as a generalizable extraction of knowledge from data. The core concept of semantic based search engine project developed by my …

A Completely Covert Audio Channel In Android, Sukanya Thakur

Master's Projects

Exfilteration of private data is a potential security threat against mobile devices. Previous research concerning such threats has generally focused on techniques that are only valid over short distances (NFC, Bluetooth, electromagnetic emanations, and so on). In this research, we develop and analyze an exfilteration attack that has no distance limitation. Specifically, we take advantage of vulnerabilities in Android that enable us to covertly record and exfilterate a voice call. This paper presents a successful implementation of our attack, which records a call (both uplink and downlink voice streams), and inaudibly transmits the recorded voice over a subsequent inaudible call, …

Sharedwealth: A Cryptocurrency To Reward Miners Evenly, Siddiq Ahmed Syed

Master's Projects

Bitcoin [19] is a decentralized cryptocurrency that has recently gained popularity and has emerged as a popular medium of exchange. The total market capitalization is around 1.5 billion US dollars as of October 2013 [28]. All the operations of Bitcoin are maintained in a distributed public global ledger known as a block chain which consists of all the successful transactions that have ever taken place. The security of a block chain is maintained by a chain of cryptographic puzzles solved by participants called miners, who in return are rewarded with bitcoins. To be successful, the miner has to put in …

Metamorphic Java Engine, Sailee Choudhary

Master's Projects

Malware is a software program outlined to damage or perform other unwanted actions to a computer system. Metamorphic malware is a category of malignant software programs that has the ability to change its code as it propagates. A hidden Markov model (HMM) is a statistical model where the system is assumed to be a Markov process with unseen states. An HMM is based on the use of statistics to detect patterns, and hence in metamorphic virus detection. Previous work has been done in order to create morphing engines using LLVM-bytecode format. This project includes the creation of a morphing engine …

Neural Network Captcha Cracker, Geetika Garg

Master's Projects

NEURAL NETWORK CAPTCHA CRACKER A CAPTCHA (acronym for "Completely Automated Public Turing test to tell Computers and Humans Apart") is a type of challenge-response test used to determine whether or not a user providing the response is human. In this project, we used a deep neural network framework for CAPTCHA recognition. The core idea of the project is to learn a model that breaks image-based CAPTCHAs. We used convolutional neural networks and recurrent neural networks instead of the conventional methods of CAPTCHA breaking based on segmenting and recognizing a CAPTCHA. Our models consist of two convolutional layers to learn image …

Entity And Relational Queries Over Big Data Storage, Nachappa Achakalera Ponnappa

Master's Projects

Big data storage involves using NoSQL technologies to handle and process huge volumes of data. NoSQL databases are non-relational, schema-free where data is stored as key-value pairs. The aim of the thesis is to implement Entity and Relational queries on top of Big Data storage. In order to achieve this, we use NoSQL technologies like MongoDB and HBase. We implement various methodologies and solutions on top of MongoDB and HBase to map data across different tables and implement entity and relational queries to retrieve entities from huge volumes of data. We also measure the performance of both the technologies and …

Comparative Analysis Of Two Clustering Algorithms: K-Means And Fsdp (Fast Search And Find Of Density Peaks), Li Miao

Master's Projects

With the overwhelming amount of data pouring into our lives, obtaining meaningful information from them is becoming a must task for people. How can people mine for "gold" in this area? Or, what tools can they use to do that? It has been proved that clustering is one of the best tools. In this project, two clustering algorithms are studied and numerically compared with various data sets. The first one is the K-means clustering which starts with initial roughly-guessed clusters, tries to classify some data points into one cluster, and iteratively repeats until converges. The second algorithm is called Fast …

Implementing Type Inference In Jedi, Vaibhav Kamble

Master's Projects

This thesis begins with an overview of type systems: evolution, concepts, and problems. This survey is based on type systems of modern languages like Scala and Haskell. Scala has a very sophisticated type system that includes generics, polymorphism, and closures. It has a built-in type inference mechanism that enables the programmer to exclude certain type annotations. It is often not required in Scala to mention the type of a variable because the compiler can infer the type from the initialization of the variable. Study of such type system is demonstrated by the implementation of the type system. A type system …

Pattern-Aided Regression Modelling And Prediction Model Analysis, Naresh Avva

Master's Projects

In this research, we develop an application for generating a pattern aided regression (PXR) model, a new type of regression model designed to represent accurate and interpretable prediction model. Our goal is to generate a PXR model using Contrast Pattern Aided Regression (CPXR) method and compare it with the multiple linear regression method. The PXR models built by CPXR are very accurate in general, often outperforming state-of-the-art regression methods by big margins. CPXR is especially effective for high-dimensional data. We use pruning to improve the classification accuracy and to remove outliers from the dataset. We provide implementation details and give …

Extensible Authentication Protocol Vulnerabilities And Improvements, Akshay Baheti

Master's Projects

Extensible Authentication Protocol(EAP) is a widely used security protocol for Wireless networks around the world. The project examines different security issues with the EAP based protocols, the family of security protocols for Wireless LAN. The project discovers an attack on the subscriber identity module(SIM) based extension of EAP. The attack is a Denial-of-Service attack that exploits the error handling mechanism in EAP protocols. The project further proposes countermeasures for detection and a defense against the discovered attack. The discovered attack can be prevented by changing the protocol to delay the processing of protocol error messages.

Cryptanalysis Of The Purple Cipher Using Random Restarts, Aparna Shikhare

Master's Projects

Cryptanalysis is the process of trying to analyze ciphers, cipher text, and crypto systems, which may exploit any loopholes or weaknesses in the systems, leading us to an understanding of the key used to encrypt the data. This project uses Expectation Maximization (EM) approach using numerous restarts to attack decipherment problems such as the Purple Cipher. In this research, we perform cryptanalysis of the Purple cipher using genetic algorithms and hidden Markov models (HMM). If the Purple cipher has a fixed plugboard, we show that genetic algorithms are successful in retrieving the plaintext from cipher text with high accuracy. On …

Pattern-Driven Programming In Scala, Huaxin Pang

Master's Projects

This is an experimental exploration of the pattern-driven programming paradigm—the sole use of pattern matching to determine the next instruction or execute. We define a pure pattern-driven programming language named PA-Scala by defining a subset of the Scala programming language, which restricts sequence control to the powerful pattern matching facilities in Scala. We use PA-Scala to explore the strengths and limitations of pattern-driven programming. By implementing a phrase structure grammar solver in PA-Scala, we show that pattern-driven programming can be used to solve general computation problems. We then implement a Prolog interpreter in PA-Scala, which demonstrates how resolution and unification …

A Javascript And Php Epub Reader Web Application, Xiaqing He

Master's Projects

EPUB is one of the most popular ebook formats. It is supported by many e-Readers, such as Apple’s iBooks, BlackBerry Playbooks, Sony Reader, Kobo eReader, Amazon Kindle Fire, and the Mozilla Firefox add-on EPUBReader. In this report, we describe our implementation of a JavaScript and PHP EPUB reader web application. This web application allows users to read the EPUB format books easily across multiple devices without any specified platform lock-in. Our application provides an EPUB library. When a user logs in, he can borrow an EPUB book from the library and save into his own bookshelf, then choose any one …

Designing A Programming Contract Library For Java, Neha Rajkumar

Master's Projects

Programmers are now developing large and complex software systems, so it’s important to have software that is consistent, efficient, and robust. Programming contracts allow developers to specify preconditions, postconditions, and invariants in order to more easily identify programming errors. The design by contract principle [1] was first used in the Eiffel programming language [2], and has since been extended to libraries in many other languages. The purpose of my project is to design a programming contract library for Java. The library supports a set of preconditions, postconditions, and invariants that are specified in Java annotations. It incorporates contract checking for …

Scalable Techniques For Similarity Search, Siddartha Reddy Nagireddy

Master's Projects

Document similarity is similar to the nearest neighbour problem and has applications in various domains. In order to determine the similarity / dissimilarity of the documents first they need to be converted into sets containing shingles. Each document is converted into k-shingles, k being the length of each shingle. The similarity is calculated using Jaccard distance between sets and output into a characteristic matrix, the complexity to parse this matrix is significantly high especially when the sets are large. In this project we explore various approaches such as Min hashing, LSH & Bloom Filter to decrease the matrix size and …

On-The-Fly Map Generator For Openstreetmap Data Using Webgl, Sreenidhi Pundi Muralidharan

Master's Projects

This project describes an approach to create an On-the-fly Map Generator for Openstreetmap Data Using WebGL. The most common methods to generate online maps generate PNG overlay tile images from a wide range of data sources, like GeoJSON, GeoTIFF, PostGIS, CSV, and SQLite, etc., based on the coordinates and zoom-level. This project aims to send vector data for the map to the browser and hence render maps on-the-fly using WebGL. We push all of the vector computation to the GPU. This means that less data needs to be sent to the browser. We have compared existing approaches to our method …

Measuring Malware Evolution, Poonkodi Ponnambalam

Master's Projects

In this research, we simulate the effect of code evolution by applying a variety of code morphing strategies. Specifically, we consider code substitution, transposition, insertion, and deletion. We then analyze the effect of these code morphing strategies relative to a variety of malware scores that have been considered in previous research. Our goal is to gain a better understanding of the strengths and weaknesses of these various malware scoring techniques. This research should prove useful in designing more robust scores for detecting malware.

Load Balancing For Entity Matching Over Big Data Using Sorted Neighborhood, Yogesh Wattamwar

Master's Projects

Entity matching also known as entity resolution, duplicate identification, reference reconciliation or record linkage and is a critically important task for data cleaning and data integration. One can think of it, as the task of finding entities matching to the same entity in the real world. These entities can belong to a single source of data, or distributed data-sources. It takes structured data as an input and process includes comparison of that structured data (entity or database record) with entities present in the knowledge base. For large-scale entity, matching data has to go through some sequence of steps, which includes …

Relationship Based Entity Recommendation System, Rakhi Poonam Verma

Master's Projects

With the increase in usage of the internet as a place to search for information, the importance of the level of relevance of the results returned by search engines have increased by many folds in recent years. In this paper, we propose techniques to improve the relevance of results shown by a search engine, by using the kinds of relationships between entities a user is interested in. We propose a technique that uses relationships between entities to recommend related entities from a knowledge base which is a collection of entities and the relationships with which they are connected to other …

A Recommendation Engine Using Apache Spark, Swapna Kulkarni

Master's Projects

The volume of structured and unstructured data has grown at exponential scale in recent days. As a result of this rapid data growth, we are always inundated with plethora of choices in any product or service. It is very natural to get lost in the amazon of such choices and finding hard to make decisions. The project aims at addressing this problem by using entity recommendation. The two main aspects that the project concentrates on are implementing and presenting more accurate entity recommendations to the user and another is dealing with vast amount of data. The project aims at presenting …

Load Balancing For Big Data Entity Matching Using Block Split, Akhilesh Kondra

Master's Projects

Entity Matching (EM) is a complex problem and has great impact on data quality. In EM we usually match all the combination of entity pairs using different similarity measures and judge if there is any match between entities. Mapreduce based parallel programing model can be used to match these entities. Even distribution of data into the map and reduce tasks will play vital role in the productivity of Mapreduce based programing model. If the dataset is large and has skewed data, then the distribution should be done effectively to achieve load balancing. In this paper, I have implemented an approach …

Bitfed, A Centralized Cryptocurrency With Distributed Miners, Shruti Sharma

Master's Projects

Bitcoin is a decentralized peer-to-peer electronic currency wherein all the payments are sent from one transactor to another directly [1]. Financial institutions are not present in the protocol, hence, there are lower processing fees. The distributed nature provides resilience to Bitcoin transactions, and it operates on mathematical principles and cryptographic proofs. As per Bitcoin generation algorithm, the number of bitcoins in existence will never surpass 21 million, which will lead to deflation and encourage hoarding. In this project, we have implemented a Bitcoin-like currency in order to mitigate the issue of deflation [7]. The idea for our protocol is based …

Metamorphic Code Generator Based On Bytecode Of Llvm Ir, Arjun Shah

Master's Projects

Metamorphic software is famous for changing the internal structure of the code while keeping the functionality same. In order to escape the signature detection along with some advanced detection techniques, many malware writers have used metamorphism as the means. On the other hand, code morphing technique increases the diversity of the software which is considered to be a potential security advantage. In our paper, we have developed a metamorphic code generator based on the LLVM framework. The architecture of LLVM has a three-phase compiler design which includes the front end, the optimizer and the back end. It also gives assistance …