Computer Sciences Commons^™

Improving Text Classification With Word Embedding, Lihao Ge

Master's Projects

One challenge in text classification is that it is hard to make feature reduction basing upon the meaning of the features. An improper feature reduction may even worsen the classification accuracy. Word2Vec, a word embedding method, has recently been gaining popularity due to its high precision rate of analyzing the semantic similarity between words at relatively low computational cost. However, there are only a limited number of researchers focusing on feature reduction using Word2Vec. In this project, we developed a Word2Vec based method to reduce the feature size while increasing the classification accuracy. The feature reduction is achieved by loosely …

Viral Marketing For Smart Cities: Influencers In Social Network Communities, Madhura Kaple, Ketki Kulkarni, Katerina Potika

Faculty Publications, Computer Science

Social networks are used by cities primarily for announcing local-area events, but also for increasing engagement of citizens in votes and elections. Given the current plethora of heterogeneous social networks, city administrators can benefit from social networks to promote initiatives, which are important to a current smart city as well use them to discover future needs in order to manage resources more efficiently. Our focus in this paper is how we can adapt commercial and viral marketing techniques to smart city systems to influence the behavior, opinion and choices of citizens in order to improve their well being and that …

Housing Price Prediction Using Support Vector Regression, Jiao Yang Wu

Master's Projects

The relationship between house prices and the economy is an important motivating factor for predicting house prices. Housing price trends are not only the concern of buyers and sellers, but it also indicates the current economic situation. Therefore, it is important to predict housing prices without bias to help both the buyers and sellers make their decisions. This project uses an open source dataset, which include 20 explanatory features and 21,613 entries of housing sales in King County, USA. We compare different feature selection methods and feature extraction algorithm with Support Vector Regression (SVR) to predict the house prices in …

An Open Source Discussion Group Recommendation System, Sarika Padmashali

Master's Projects

A recommendation system analyzes user behavior on a website to make suggestions about what a user should do in the future on the website. It basically tries to predict the “rating” or “preference” a user would have for an action. Yioop is an open source search engine, wiki system, and user discussion group system managed by Dr. Christopher Pollett at SJSU. In this project, we have developed a recommendation system for Yioop where users are given suggestions about the threads and groups they could join based on their user history. We have used collaborative filtering techniques to make recommendations and …

Neural Net Stock Trend Predictor, Sonal Kabra

Master's Projects

This report analyzes new and existing stock market prediction techniques. Traditional technical analysis was combined with various machine-learning approaches such as artificial neural networks, k-nearest neighbors, and decision trees. Experiments we conducted show that technical analysis together with machine learning can be used to profitably direct an investor’s trading decisions. We are measuring the profitability of experiments by calculating the percentage weekly return for each stock entity under study. Our algorithms and simulations are developed using Python. The technical analysis methodology combined with machine learning algorithms show promising results which we discuss in this report.

Predicting Pancreatic Cancer Using Support Vector Machine, Akshay Bodkhe

Master's Projects

This report presents an approach to predict pancreatic cancer using Support Vector Machine Classification algorithm. The research objective of this project it to predict pancreatic cancer on just genomic, just clinical and combination of genomic and clinical data. We have used real genomic data having 22,763 samples and 154 features per sample. We have also created Synthetic Clinical data having 400 samples and 7 features per sample in order to predict accuracy of just clinical data. To validate the hypothesis, we have combined synthetic clinical data with subset of features from real genomic data. In our results, we observed that …

Adding Differential Privacy In An Open Board Discussion Board System, Pragya Rana

Master's Projects

This project implements a privacy system for statistics generated by the Yioop search and discussion board system. Statistical data for such a system consists of various counts, sums, and averages that might be displayed for groups, threads, etc. When statistical data is made publicly available, there is no guarantee of preserving the privacy of an individual. Ideally, any data extracted should not reveal any sensitive information about an individual. In order to help achieve this, we implemented a Differential Privacy mechanism for Yioop. Differential privacy preserves privacy up to some controllable parameters of the number of items or individuals being …

Path-Finding Methodology For Visually-Impaired Patients Based On Image-Processing, Abhilash Goyal

Master's Projects

The objective of this project is to propose and develop the path-finding methodology for the visually impaired patients. The proposed novel methodology is based on image-processing and it is targeted for the patients who are not completely blind. The major problem faced by visually impaired patients is to walk independently. It is mainly because these patients can not see obstacles in front of them due to the degradation in their eye sight. Degradation in the eye-sight is mainly because either the light doesn't focus on the retina properly or due to the malfunction of the photoreceptor cells on the retina, …

Credit Scoring Using Logistic Regression, Ansen Mathew

Master's Projects

This report presents an approach to predict the credit scores of customers using the Logistic Regression machine learning algorithm. The research objective of this project is to perform a comparative study between feature selection and feature extraction, against the same dataset using the Logistic Regression machine learning algorithm. For feature selection, we have used Stepwise Logistic Regression. For feature extraction, we have used Singular Value Decomposition (SVD) and Weighted Singular Value Decomposition (SVD). In order to test the accuracy obtained using feature selection and feature extraction, we used a public credit dataset having 11 features and 150,000 records. After performing …

Web - Based Office Market, Manodivya Kathiravan

Master's Projects

People who work in an office often have different pools of resources that they want to exchange. They want to trade their resources/work(seller) with a person who wants that particular resource(buyer) and in return get another resource the buyer offers. These kind of exchanges are often called Barter-exchanges where an item is traded for another item without the involvement of actual money. An exchange is set to be complete when there is a match between an available item and a desired item. This exchange is called direct exchange. When an item desired by one user is made available through a …

Ai For Classic Video Games Using Reinforcement Learning, Shivika Sodhi

Master's Projects

Deep reinforcement learning is a technique to teach machines tasks based on trial and error experiences in the way humans learn. In this paper, some preliminary research is done to understand how reinforcement learning and deep learning techniques can be combined to train an agent to play Archon, a classic video game. We compare two methods to estimate a Q function, the function used to compute the best action to take at each point in the game. In the first approach, we used a Q table to store the states and weights of the corresponding actions. In our experiments, this …

Document Classification Using Machine Learning, Ankit Basarkar

Master's Projects

To perform document classification algorithmically, documents need to be represented such that it is understandable to the machine learning classifier. The report discusses the different types of feature vectors through which document can be represented and later classified. The project aims at comparing the Binary, Count and TfIdf feature vectors and their impact on document classification. To test how well each of the three mentioned feature vectors perform, we used the 20-newsgroup dataset and converted the documents to all the three feature vectors. For each feature vector representation, we trained the Naïve Bayes classifier and then tested the generated classifier …

Community Detection In Social Networks, Ketki Kulkarni

Master's Projects

The rise of the Internet has brought people closer. The number of interactions between people across the globe has gone substantially up due to social awareness, the advancements of the technology, and digital interaction. Social networking sites have built societies, communities virtually. Often these societies are displayed as a network of nodes depicting people and edges depicting relationships, links. This is a good and e cient way to store, model and represent systems which have a complex and rich information. Towards that goal we need to nd e ective, quick methods to analyze social networks. One of the possible solution …

Reducing Query Latency For Information Retrieval, Swapnil Satish Kamble

Master's Projects

As the world is moving towards Big Data, NoSQL (Not only SQL) databases are gaining much more popularity. Among the other advantages of NoSQL databases, one of their key advantage is that they facilitate faster retrieval for huge volumes of data, as compared to traditional relational databases. This project deals with one such popular NoSQL database, Apache HBase. It performs quite efficiently in cases of retrieving information using the rowkey (similar to a primary key in a SQL database). But, in cases where one needs to get information based on non-rowkey columns, the response latency is higher than what we …

Cascaded Facial Detection Algorithms To Improve Recognition, Edmund Yee

Master's Projects

The desire to be able to use computer programs to recognize certain biometric qualities of people have been desired by several different types of organizations. One of these qualities worked on and has achieved moderate success is facial detection and recognition. Being able to use computers to determine where and who a face is has generated several different algorithms to solve this problem with different benefits and drawbacks. At the backbone of each algorithm is the desire for it to be quick and accurate. By cascading face detection algorithms, accuracy can be improved but runtime will subsequently be increased. Neural …

Library For Writing Contracts For Java Programs Using Prolog, Yogesh Dixit

Master's Projects

Today many large and complex software systems are being developed in Java. Although, software always has bugs, it is very important that these developed systems are more reliable despite these bugs.

One way that we can help achieve this is the Design by Contract (DbC) paradigm, which was first introduced by Bertrand Meyer, the creator of Eiffel. The concept of DbC was introduced for software developers so that they can produce more reliable software systems with a little extra cost. Using programming contracts allows developer to specify details such as input conditions and expected output conditions. Doing this makes it …

Dynamic Information Flow Analysis In Ruby, Vigneshwari Chandrasekaran

Master's Projects

With the rapid increase in usage of the internet and online applications, there is a huge demand for applications to handle data privacy and integrity. Applications are already complex with business logic; adding the data safety logic would make them more complicated. The more complex the code becomes, the more possibilities it opens for security-critical bugs. To solve this conundrum, we can push this data safety handling feature to the language level rather than the application level. With a secure language, developers can write their application without having to worry about data security.

This project introduces dynamic information flow analysis …

Computational Analysis Of Cryptic Splice Sites, Remya Mohanan

Master's Projects

DNA in the nucleus of all eukaryotes is transcribed into mRNA where it is then translated into proteins. The DNA which is transcribed into mRNA is composed of coding and non-coding regions called exons and introns, respectively. It undergoes a post-trancriptional process called splicing where the introns or the non-coding regions are removed from the pre-mRNA to give the mature mRNA. Splicing of pre-mRNAs at 5 ́ and 3ˊ ends is a crucial step in the gene expression pathway. The mis-splicing by the spliceosome at different sites known as cryptic splice sites is caused by mutations which will affect the …

Influence Detection And Spread Estimation In Social Networks, Madhura Kaple

Master's Projects

A social network is an online platform, where people communicate and share information with each other. Popular social network features, which make them di erent from traditional communication platforms, are: following a user, re-tweeting a post, liking and commenting on a post etc. Many companies use various social networking platforms extensively as a medium for marketing their products. A xed amount of budget is alloted by the companies to maximize the positive in uence of their product. Every social network consists of a set of users (people) with connections between them. Each user has the potential to extend its in …

Comparing Authentic And Cryptic 5’ Splice Sites Using Hidden Markov Models And Decision Trees, Pratikshya Mishra

Master's Projects

Splicing is the editing of the precursor mRNA produced during transcription. The mRNA contains a large number of nucleotides in the introns and exons which are spliced to remove the introns and bind the exons to produce the mature mRNA which is translated to generate proteins. Hence accurate splicing at 5’ and 3’ splice sites (authentic splice sites (AuthSS)) is of foremost importance. The 5’ and 3’ splice sites are characterized by consensus sequences. Eukaryotic genome also contains splice sites known as Cryptic Splice Sites (CSS) that match the consensus. But the CSS are activated only when there is a …

A Chatbot Framework For Yioop, Harika Nukala

Master's Projects

Over the past few years, messaging applications have become more popular than Social networking sites. Instead of using a specific application or website to access some service, chatbots are created on messaging platforms to allow users to interact with companies’ products and also give assistance as needed. In this project, we designed and implemented a chatbot Framework for Yioop. The goal of the Chatbot Framework for Yioop project is to provide a platform for developers in Yioop to build and deploy chatbot applications. A chatbot is a web service that can converse with users using artificial intelligence in messaging platforms. …

Shopbot: An Image Based Search Application For E-Commerce Domain, Nishant Goel

Master's Projects

For the past few years, e-commerce has changed the way people buy and sell products. People use this business model to do business over the Internet. In this domain, Human-Computer Interaction has been gaining momentum. Lately, there has been an upsurge in agent based applications in the form of intelligent personal assistants (also known as Chatbots) which make it easier for users to interact with digital services via a conversation, in the same way we talk to humans. In e- commerce, these assistants offer mainly text-based or speech based search capabilities. They can handle search for most products, but cannot …

Black Box Analysis Of Android Malware Detectors, Guruswamy Nellaivadivelu

Master's Projects

Code obfuscation can make it challenging to detect malware in Android devices. Malware writers obfuscate the code of their programs by employing various techniques that attempt to hide the true purpose of the program. Malware detectors can use a number of features to classify a program as a malware. If the malware detector uses a feature that is obfuscated, then the malware detector will likely fail to classify the malware as malicious software. In this research, we obfuscate selected features of known malware and determine whether the malware can still be detected by a given detector. Using this approach, we …

Mining Frequency Of Drug Side Effects Over A Large Twitter Dataset Using Apache Spark, Dennis Hsu

Master's Projects

Despite clinical trials by pharmaceutical companies as well as current FDA reporting systems, there are still drug side effects that have not been caught. To find a larger sample of reports, a possible way is to mine online social media. With its current widespread use, social media such as Twitter has given rise to massive amounts of data, which can be used as reports for drug side effects. To process these large datasets, Apache Spark has become popular for fast, distributed batch processing. In this work, we have improved on previous pipelines in sentimental analysis-based mining, processing, and extracting tweets …

Malware Scores Based On Image Processing, Vikash Raja Samuel Selvin

Master's Projects

Malware analysis can be based on static or dynamic analysis. Static analysis includes signature-based detection and other forms of analysis rely only on features that can be extracted without code execution or emulation. In contrast, dynamic analysis depends on features extracted at runtime (or via emulation) such as API calls, patterns of memory access, and so on. Dynamic analysis can be more informative and is generally more robust, but static analysis is typically more efficient. In this research, we implement, test, and analyze malware scores based on image processing. Previous work has shown that useful malware scores can be obtained …

Generic Online Learning For Partial Visible & Dynamic Environment With Delayed Feedback, Behrooz Shahriari

Master's Projects

Reinforcement learning (RL) has been applied to robotics and many other domains which a system must learn in real-time and interact with a dynamic environment. In most studies the state- action space that is the key part of RL is predefined. Integration of RL with deep learning method has however taken a tremendous leap forward to solve novel challenging problems such as mastering a board game of Go. The surrounding environment to the agent may not be fully visible, the environment can change over time, and the feedbacks that agent receives for its actions can have a fluctuating delay. In …

Design Concept For A Failover Mechanism In Distributed Sdn Controllers, Nathan Kong

Master's Projects

Software defined networking allows the separation of the control plane and data plane in networking. It provides scalability, programmability, and centralized control. It will use these traits to reach ubiquitous connectivity. Like all concepts software defined networking does not offer these advantages without a cost. By utilizing a centralized controller, a single point of failure is created. To address this issue, this paper proposes a distributed controller failover. This failover will provide a mechanism for recovery when controllers are not located in the same location. This failover mechanism is based on number of hops from orphan nodes to the controller …

Masquerade Detection On Mobile Devices, Swathi Nambiar Kadala Manikoth

Master's Projects

A masquerade is an attack where the attacker avoids detection by impersonating an authorized user of a system. In this research we consider the problem of masquerade detection on mobile devices. Our goal is to improve on previous work by considering more features and a wide variety of machine learning techniques. Our approach consists of verifying the authenticity of users based on individual features and combinations of features for all users to determine which features contribute the most to masquerade detection. Also, we determine which of the two approaches - the combination of features or using individual features has performed …

Transcriptase–Light: A Polymorphic Virus Construction Kit, Saurabh Borwankar

Master's Projects

Many websites use JavaScript to display dynamic and interactive content. Hence, attackers are developing JavaScript–based malware. In this paper, we focus on Transcriptase JavaScript malware.

The high–level and dynamic nature of the JavaScript language helps malware writers to create polymorphic and metamorphic malware using obfuscation techniques. These types of malware change their internal structure on each infection, making them difficult to detect with traditional methods. These types of malware can be detected using machine learning methods.

This project creates Transcriptase–Light, a new polymorphic construction kit. We perform an experiment with the Transcriptase–Light against a hidden Markov model. Our experiment shows …

Switching Between Page Replacement Algorithms Based On Work Load During Runtime In Linux Kernel, Praveen Subramaniyam

Master's Projects

Today’s computers are equipped with multiple processor cores to execute multiple programs effectively at a single point of time. This increase in the number of cores needs to be equipped with a huge amount of physical memory to keep multiple applications in memory at a time and to effectively switch between them, without getting affected by the low speed disk memory. The physical memory of today’s world has become so cheap such that all the computer systems are always equipped with sufficient amount of physical memory required effectively to run most of the applications. Along with the memory, the sizes …