Digital Commons Network^™

Effects Of Training Datasets On Both The Extreme Learning Machine And Support Vector Machine For Target Audience Identification On Twitter, Siaw Ling Lo, David Cornforth, Raymond Chiong

Research Collection School Of Computing and Information Systems

The ability to identify or predict a target audience from the increasingly crowded social space will provide a company some competitive advantage over other companies. In this paper, we analyze various training datasets, which include Twitter contents of an account owner and its list of followers, using features generated in different ways for two machine learning approaches - the Extreme Learning Machine (ELM) and Support Vector Machine (SVM). Various configurations of the ELM and SVM have been evaluated. The results indicate that training datasets using features generated from the owner tweets achieve the best performance, relative to other feature sets. …

Understanding Online Fashion Networks, Kota Yamaguchi

Stony Brook Theses and Dissertations Collection, 2006-2020 (closed to submissions)

Emergence of online social networks has transformed how people interact with digital media. Any user is a consumer and a publisher of media content, such as texts, images, or videos, in the online community. In such an environment, it is crucial to develop technology to help people by organizing and utilizing plethora of media content to meet the community demands. To this goal, this dissertation studies and tries to establish computational approaches to understand networked content, using large-scale data from a real-world online fashion network, Chictopia. This study focuses on two major challenges of fashion networks: understanding of visual content, …

Avatar Captcha : Telling Computers And Humans Apart Via Face Classification And Mouse Dynamics., Darryl Felix D’Souza

Electronic Theses and Dissertations

Bots are malicious, automated computer programs that execute malicious scripts and predefined functions on an affected computer. They pose cybersecurity threats and are one of the most sophisticated and common types of cybercrime tools today. They spread viruses, generate spam, steal personal sensitive information, rig online polls and commit other types of online crime and fraud. They sneak into unprotected systems through the Internet by seeking vulnerable entry points. They access the system’s resources like a human user does. Now the question arises how do we counter this? How do we prevent bots and on the other hand allow human …

Temporal Contextual Descriptors And Applications To Emotion Analysis., Haythem Balti

Electronic Theses and Dissertations

The current trends in technology suggest that the next generation of services and devices allows smarter customization and automatic context recognition. Computers learn the behavior of the users and can offer them customized services depending on the context, location, and preferences. One of the most important challenges in human-machine interaction is the proper understanding of human emotions by machines and automated systems. In the recent years, the progress made in machine learning and pattern recognition led to the development of algorithms that are able to learn the detection and identification of human emotions from experience. These algorithms use different modalities …

Multilingual Named Entity Recognition, Vivek V. Kulkarni

Stony Brook Theses and Dissertations Collection, 2006-2020 (closed to submissions)

With the massive amounts of unannotated text available from myriad sources, learning representations useful for natural language processing(NLP) tasks is an increasingly popular research area. Using deep learning techniques, we learn distributed representations for words (word embeddings) using Wikipedia as the source of text for 40 languages. These distributed representations represent each word as a point in feature space and capture useful semantic and syntactic properties of words amd have been shown to be useful in NLP Tasks like Part of Speech Tagging(POS) etc. We have built 2 classes of word embeddings namely Polyglot and Skipgram for these languages. We …

Pcp-Ml: Protein Characterization Package For Machine Learning, Jesse Eickholt, Zheng Wang

Faculty Publications

Background: Machine Learning (ML) has a number of demonstrated applications in protein prediction tasks such as protein structure prediction. To speed further development of machine learning based tools and their release to the community, we have developed a package which characterizes several aspects of a protein commonly used for protein prediction tasks with machine learning.

Findings: A number of software libraries and modules exist for handling protein related data. The package we present in this work, PCP-ML, is unique in its small footprint and emphasis on machine learning. Its primary focus is on characterizing various aspects of a …

Identifying The High-Value Social Audience From Twitter Through Text-Mining Methods, Siaw Ling Lo, David Cornforth, Raymond Chiong

Research Collection School Of Computing and Information Systems

Doing business on social media has become a common practice for many companies these days. While the contents shared on Twitter and Facebook offer plenty of opportunities to uncover business insights, it remains a challenge to sift through the huge amount of social media data and identify the potential social audience who is highly likely to be interested in a particular company. In this paper, we analyze the Twitter content of an account owner and its list of followers through various text mining methods, which include fuzzy keyword matching, statistical topic modeling and machine learning approaches. We use tweets of …

An Automatic Similarity Detection Engine Between Sacred Texts Using Text Mining And Similarity Measures, Salha Hassan Muhammed Qahl

Theses

Is there any similarity between the contexts of the Holy Bible and the Holy Quran, and can this be proven mathematically? The purpose of this research is using the Bible and the Quran as our corpus, we explore the performance of various feature extraction and machine learning techniques. The unstructured nature of text data adds an extra layer of complexity in the feature extraction task, and the inherently sparse nature of the corresponding data matrices makes text mining a distinctly difficult task. Among other things, We assess the difference between domain-based syntactic feature extraction and domain-free feature extraction, and then …

Survey On Fall Detection And Fall Prevention Using Wearable And External Sensors, Yueng Santiago Delahoz, Miguel Angel Labrador

Computer Science and Engineering Faculty Publications

According to nihseniorhealth.gov (a website for older adults), falling represents a great threat as people get older, and providing mechanisms to detect and prevent falls is critical to improve people’s lives. Over 1.6 million U.S. adults are treated for fall-related injuries in emergency rooms every year suffering fractures, loss of independence, and even death. It is clear then, that this problem must be addressed in a prompt manner, and the use of pervasive computing plays a key role to achieve this. Fall detection (FD) and fall prevention (FP) are research areas that have been active for over a decade, and …

Training Set Design For Test Removal Classication In Ic Test, Nagarjun Hassan Ranganath

Dissertations and Theses

This thesis reports the performance of a simple classifier as a function of its training data set. The classifier is used to remove analog tests and is named the Test Removal Classifier (TRC).

The thesis proposes seven different training data set designs that vary by the number of wafers in the data set, the source of the wafers and the replacement scheme of the wafers. The training data set size ranges from a single wafer to a maximum of five wafers. Three of the training data sets include wafers from the Lot Under Test (LUT). The training wafers in the …

Supervised Material Classification In Oblique Aerial Imagery Using Gabor Filter Features, Michael L. Harris

Theses

RIT's Digital Imaging and Remote Sensing Image Generation (DIRSIG) tool allows modeling of real world scenes to create synthetic imagery for sensor design and analysis, trade studies, algorithm validation, and training image analysts. To increase model construction speed, and the diversity and size of synthetic scenes which can be generated it is desirable to automatically segment real world imagery into different material types and import a material classmap into DIRSIG.

This work contributes a methodology based on standard texture recognition techniques to supervised classification of material types in oblique aerial imagery. Oblique imagery provides many challenges for texture recognition due …

Wideband Spectrum Sensing And Signal Classification For Autonomous Self-Learning Cognitive Radios, Mario Bkassiny

Electrical and Computer Engineering ETDs

In this dissertation, we develop a novel cognitive radio (CR) architecture, referred to as the Radiobot [1], whose goals go beyond dynamic spectrum access (DSA) to achieve the main features of cognition, notably, self-learning and self-reconfiguration. The proposed CR architecture is based on a sequence of signal processing and machine learning techniques that enable the Radiobot to sense a wide frequency band and act autonomously by learning from past experience. To achieve its goals, the proposed CR is equipped with the following functionalities: 1) Wideband spectrum sensing, 2) non-parametric signal classification, 3) unsupervised learning and reasoning and 4) decentralized decision-making. …

Intelligent Indexing: A Semi-Automated, Trainable System For Field Labeling, Robert T. Clawson

Theses and Dissertations

We present Intelligent Indexing: a general, scalable, collaborative approach to indexing and transcription of non-machine-readable documents that exploits visual consensus and group labeling while harnessing human recognition and domain expertise. In our system, indexers work directly on the page, and with minimal context switching can navigate the page, enter labels, and interact with the recognition engine. Interaction with the recognition engine occurs through preview windows that allow the indexer to quickly verify and correct recommendations. This interaction is far superior to conventional, tedious, inefficient post-correction and editing. Intelligent Indexing is a trainable system that improves over time and can provide …

A Brain Computer Interface For Interactive And Intelligent Image Search And Retrieval, Shitij P. Kumar

Theses

This research proposes a Brain Computer Interface as an interactive and intelligent Image Search and Retrieval tool that allows users, disabled or otherwise to browse and search for images using brain signals. The proposed BCI system implements decoding the brain state by using a non-invasive electroencephalography (EEG) signals, in combination with machine learning, artificial intelligence and automatic content and similarity analysis of images. The user can spell search queries using a mental typewriter (Hex-O-Speller), and the resulting images from the web search are shown to the user as a Rapid Serial Visual Presentations (RSVP). For each image shown, the EEG …

Model-Free Method Of Reinforcement Learning For Visual Tasks, Jeff S. Soldate, Jonghoon Jin, Eugenio Culurciello

The Summer Undergraduate Research Fellowship (SURF) Symposium

There has been success in recent years for neural networks in applications requiring high level intelligence such as categorization and assessment. In this work, we present a neural network model to learn control policies using reinforcement learning. It takes a raw pixel representation of the current state and outputs an approximation of a Q value function made with a neural network that represents the expected reward for each possible state-action pair. The action is chosen an \epsilon-greedy policy, choosing the highest expected reward with a small chance of random action. We used gradient descent to update the weights and biases …

Collaborative Online Multitask Learning, Guangxia Li, Steven C. H. Hoi, Kuiyu Chang, Wenting Liu, Ramesh Jain

Research Collection School Of Computing and Information Systems

We study the problem of online multitask learning for solving multiple related classification tasks in parallel, aiming at classifying every sequence of data received by each task accurately and efficiently. One practical example of online multitask learning is the micro-blog sentiment detection on a group of users, which classifies micro-blog posts generated by each user into emotional or non-emotional categories. This particular online learning task is challenging for a number of reasons. First of all, to meet the critical requirements of online applications, a highly efficient and scalable classification solution that can make immediate predictions with low learning cost is …

Improving Structural Features Prediction In Protein Structure Modeling, Ashraf Yaseen

Computer Science Theses & Dissertations

Proteins play a vital role in the biological activities of all living species. In nature, a protein folds into a specific and energetically favorable three-dimensional structure which is critical to its biological function. Hence, there has been a great effort by researchers in both experimentally determining and computationally predicting the structures of proteins.

The current experimental methods of protein structure determination are complicated, time-consuming, and expensive. On the other hand, the sequencing of proteins is fast, simple, and relatively less expensive. Thus, the gap between the number of known sequences and the determined structures is growing, and is expected to …

Scalable Automation Of Online Network Attack Characterization, Ryan T. Rawlins

Theses

Cyber attacks to enterprise networks and critical infrastructures are becoming more prevalent and diverse. Timely recognition of attack strategies and behaviors will assist analysts or resilient network defense systems in deploying effective means in anticipation of future threats. An attack can be characterized by the sequences of observed events that are relevant to critical assets. Earlier work has developed a semi-supervised learning framework to process large-scale events and extract attack behaviors. While the framework is designed to support online processing, the implementation requires extension and restructuring to support scalable automation of sustainable online network attack characterization.

This work builds upon …

Towards An Automated Weight Lifting Coach: Introducing Lift, Michael Andrew Lady

Master's Theses

The fitness device market is young and rapidly growing. More people than ever before take count of how many steps they walk, how many calories they burn, their heart rate over time, and even their quality of sleep. New, and as of yet, unreleased fitness devices have promised the next evolution of functionality with exercise technique analysis. These next generation of fitness devices have wrist and armband style form factors, which may not be optimal for barbell exercises such as back squat, bench press, and overhead press where a sensor on one arm may not provide the most relevant data …

Energy-Efficient Stdp-Based Learning Circuits With Memristor Synapses, Xinyu Wu, Vishal Saxena, Kristy A. Campbell

Electrical and Computer Engineering Faculty Publications and Presentations

It is now accepted that the traditional von Neumann architecture, with processor and memory separation, is ill suited to process parallel data streams which a mammalian brain can efficiently handle. Moreover, researchers now envision computing architectures which enable cognitive processing of massive amounts of data by identifying spatio-temporal relationships in real-time and solving complex pattern recognition problems. Memristor cross-point arrays, integrated with standard CMOS technology, are expected to result in massively parallel and low-power Neuromorphic computing architectures. Recently, significant progress has been made in spiking neural networks (SNN) which emulate data processing in the cortical brain. These architectures comprise of …

Structural Health Monitoring Of An In-Service Highway Bridge With Uncertainties, Jingcheng Li

Doctoral Dissertations

In recent years, structural health monitoring (SHM) has become increasingly important for the assessment and maintenance of aging infrastructure. However, some challenges remain in applying SHM in the field. For example, high cost of wired sensors limits the SHM application to small number of structures within a network. Wireless smart sensors (WSSs), on the other hand, provide a versatile, cost-effective method of monitoring dynamic conditions of structures in near real-time. While the wireless sensor technology has been available for nearly a decade, there have been limited numbers of full-scale applications due to its intrinsic limitations such as power supply issue. …

High Performance Implementation Of Support Vector Machines Using Opencl, Ethan Peters

Theses

Support Vector Machines are a machine learning approach that is well studied, thoroughly vetted and effective in a large number of applications. The objective of this thesis is to accelerate an implementation of Support Vector Machines (SVM) using a heterogeneous computing system programmed using OpenCL in C/C++. LIBSVM, a widely-available, popular and open source implementation of SVM is chosen, allowing the presented work to be integrated seamlessly into existing systems. The proposed framework is evaluated in terms of speed and accuracy when performing training and classification on a number of standard data sets. Testing was based on two work station …

Ensemble Methods For Historical Machine-Printed Document Recognition, William B. Lund

Theses and Dissertations

The usefulness of digitized documents is directly related to the quality of the extracted text. Optical Character Recognition (OCR) has reached a point where well-formatted and clean machine- printed documents are easily recognizable by current commercial OCR products; however, older or degraded machine-printed documents present problems to OCR engines resulting in word error rates (WER) that severely limit either automated or manual use of the extracted text. Major archives of historical machine-printed documents are being assembled around the globe, requiring an accurate transcription of the text for the automated creation of descriptive metadata, full-text searching, and information extraction. Given document …

Moving Object Detection For Interception By A Humanoid Robot, Saltanat B. Tazhibayeva

Open Access Theses

Interception of a moving object with an autonomous robot is an important problem in robotics. It has various application areas, such as in an industrial setting where products on a conveyor would be picked up by a robotic arm, in the military to halt intruders, in robotic soccer (where the robots try to get to the moving ball and try to block an opponent's attempt to pass the ball), and in other challenging situations. Interception, in and of itself, is a complex task that demands a system with target recognition capability, proper navigation and actuation toward the moving target. There …

Machine Learning In Wireless Sensor Networks: Algorithms, Strategies, And Applications, Mohammad Abu Alsheikh, Shaowei Lin, Dusit Niyato, Hwee-Pink Tan

Research Collection School Of Computing and Information Systems

Wireless sensor networks (WSNs) monitor dynamic environments that change rapidly over time. This dynamic behavior is either caused by external factors or initiated by the system designers themselves. To adapt to such conditions, sensor networks often adopt machine learning techniques to eliminate the need for unnecessary redesign. Machine learning also inspires many practical solutions that maximize resource utilization and prolong the lifespan of the network. In this paper, we present an extensive literature review over the period 2002-2013 of machine learning methods that were used to address common issues in WSNs. The advantages and disadvantages of each proposed algorithm are …

Document Classification In Support Of Automated Metadata Extraction Form Heterogeneous Collections, Paul K. Flynn

Computer Science Theses & Dissertations

A number of federal agencies, universities, laboratories, and companies are placing their documents online and making them searchable via metadata fields such as author, title, and publishing organization. To enable this, every document in the collection must be catalogued using the metadata fields. Though time consuming, the task of identifying metadata fields by inspecting the document is easy for a human. The visual cues in the formatting of the document along with accumulated knowledge and intelligence make it easy for a human to identify various metadata fields. Even with the best possible automated procedures, numerous sources of error exist, including …

Using The K-Means Clustering Algorithm To Classify Features For Choropleth Maps, Mark Polczynski, Michael Polczynski

Electrical and Computer Engineering Faculty Research and Publications

Common methods for classifying choropleth map features typically form classes based on a single feature attribute. This technical note reviews the use of the k-means clustering algorithm to perform feature classification using multiple feature attributes. The k-means clustering algorithm is described and compared to other common classification methods, and two examples of choropleth maps prepared using k-means clustering are provided.

Automatic Readability Detection For Modern Standard Arabic, Jonathan Neil Forsyth

Theses and Dissertations

Research for automatic readability prediction of text has increased in the last decade and has shown that various machine learning methods can effectively address this problem. Many researchers have applied machine learning to readability prediction for English, while Modern Standard Arabic (MSA) has received little attention. Here I describe a system which leverages machine learning to automatically predict the readability of MSA. I gathered a corpus comprising 179 documents that were annotated with the Interagency Language Roundtable (ILR) levels. Then, I extracted lexical and discourse features from each document. Finally, I applied the Tilburg Memory-Based Learning (TiMBL) machine learning system …

Retrieval-Based Face Annotation By Weak Label Regularized Local Coordinate Coding, Dayong Wang, Steven C. H. Hoi, Ying He, Jianke Zhu, Mei Tao, Jiebo Luo

Research Collection School Of Computing and Information Systems

Auto face annotation, which aims to detect human faces from a facial image and assign them proper human names, is a fundamental research problem and beneficial to many real-world applications. In this work, we address this problem by investigating a retrieval-based annotation scheme of mining massive web facial images that are freely available over the Internet. In particular, given a facial image, we first retrieve the top n similar instances from a large-scale web facial image database using content-based image retrieval techniques, and then use their labels for auto annotation. Such a scheme has two major challenges: 1) how to …

Classifying Lower Extremity Muscle Fatigue During Walking Using Machine Learning And Inertial Sensors, Jian Zhang, Thurmon Lockhart, Rahul Soangra

Physical Therapy Faculty Articles and Research

Fatigue in lower extremity musculature is associated with decline in postural stability, motor performance and alters normal walking patterns in human subjects. Automated recognition of lower extremity muscle fatigue condition may be advantageous in early detection of fall and injury risks. Supervised machine learning methods such as Support Vector Machines (SVM) have been previously used for classifying healthy and pathological gait patterns and also for separating old and young gait patterns. In this study we explore the classification potential of SVM in recognition of gait patterns utilizing an inertial measurement unit associated with lower extremity muscular fatigue. Both kinematic and …