Digital Commons Network^™

Assessing The Language Of Chat For Teamwork Dialogue, Antonette Shibani, Elizabeth Koh, Vivian Lai, Kyong Jin Shim

Research Collection School Of Computing and Information Systems

In technology enhanced language learning, many pedagogical activities involve students in online discussion such as synchronous chat, in order to help them practice their language skills. Besides developing the language competency of students, it is also crucial to nurture their teamwork competencies for today's global and complex environment. Language communication is an important glue of teamwork. In order to assess the language of chat for teamwork dimensions, several text mining methods are pos sible. However, difficulties arise such as pre-processing being a black box and classification approaches and algorithms being dependent on the context. To address these issues, the study …

Clustering Classes In Packages For Program Comprehension, Xiaobing Sun, Xiangyue Liu, Bin Li, Bixin Li, David Lo, Lingzhi Liao

Research Collection School Of Computing and Information Systems

During software maintenance and evolution, one of the important tasks faced by developers is to understand a system quickly and accurately. With the increasing size and complexity of an evolving system, program comprehension becomes an increasingly difficult activity. Given a target system for comprehension, developers may first focus on the package comprehension. The packages in the system are of different sizes. For small-sized packages in the system, developers can easily comprehend them. However, for large-sized packages, they are difficult to understand. In this article, we focus on understanding these large-sized packages and propose a novel program comprehension approach for large-sized …

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 …

Knowing How: A Computational Approach, Joseph A. Roman

Student Publications

With advances in Artificial Intelligences being achieved through the use of Artificial Neural Networks, we are now at the point where computers are able to do tasks that were previously only able to be accomplished by humans. These advancements must cause us to reconsider our previous understanding of how people come to know how to do a particular task. In order to unpack this question, I will first look to an account of knowing how presented by Jason Stanley in his book Know How. I will then look towards criticisms of this view before using evidence presented by the existence …

Aspect Extraction From Product Reviews Using Category Hierarchy Information, Yifeng Yang, Chen Cen, Minghui Qiu, Forrest Sheng Bao

Research Collection School Of Computing and Information Systems

Aspect extraction is a task to abstract the common properties of objects from corpora discussing them, such as reviews of products. Recent work on aspect extraction is leveraging the hierarchical relationship between products and their categories. However, such effort focuses on the aspects of child categories but ignores those from parent categories. Hence, we propose an LDA-based generative topic model inducing the two-layer categorical information (CAT-LDA), to balance the aspects of both a parent category and its child categories. Our hypothesis is that child categories inherit aspects from parent categories, controlled by the hierarchy between them. Experimental results on 5 …

Language Relations On Twitter: A Network Science Approach, Priya Saha

Theses and Dissertations

Online social networks (e.g., Twitter and Facebook) play a vital role in the spreading of information in today’s world. Interestingly, the spread of information is enabled by the existence of an underlying connectivity of the users. One factor influencing the online connectivity, which only recently has been receiving attention, is the language used by the user in his or her activities. The understanding of information propagation from the perspective of languages is of particular interest because we live in a world with a very diverse set of languages. Using Network Science approaches, we demonstrate that Twitter users have a strong …

Statistical Learning Methods For Mining Marketing And Biological Data, Jie Zhang

Dissertations

Nowadays, the value of data has been broadly recognized and emphasized. More and more decisions are made based on data and analysis rather than solely on experience and intuition. With the fast development of networking, data storage, and data collection capacity, data have increased dramatically in industry, science and engineering domains, which brings both great opportunities and challenges. To take advantage of the data flood, new computational methods are in demand to process, analyze and understand these datasets.

This dissertation focuses on the development of statistical learning methods for online advertising and bioinformatics to model real world data with temporal …

A Compare-Aggregate Model For Matching Text Sequences, Shuohang Wang, Jing Jiang

Research Collection School Of Computing and Information Systems

Many NLP tasks including machine comprehension, answer selection and text entailment require the comparison between sequences. Matching the important units between sequences is a key to solve these problems. In this paper, we present a general "compare-aggregate" framework that performs word-level matching followed by aggregation using Convolutional Neural Networks. We particularly focus on the different comparison functions we can use to match two vectors. We use four different datasets to evaluate the model. We find that some simple comparison functions based on element-wise operations can work better than standard neural network and neural tensor network.

Machine Comprehension Using Match-Lstm And Answer Pointer, Shuohang Wang, Jing Jiang

Research Collection School Of Computing and Information Systems

Machine comprehension of text is an important problem in natural language processing. A recently released dataset, the Stanford Question Answering Dataset (SQuAD), offers a large number of real questions and their answers created by humans through crowdsourcing. SQuAD provides a challenging testbed for evaluating machine comprehension algorithms, partly because compared with previous datasets, in SQuAD the answers do not come from a small set of candidate answers and they have variable lengths. We propose an end-to-end neural architecture for the task. The architecture is based on match-LSTM, a model we proposed previously for textual entailment, and Pointer Net, a sequence-to-sequence …

Mixed-Initiative Personal Assistants, Joshua W. Buck, Saverio Perugini

Computer Science Faculty Publications

Specification and implementation of flexible human-computer dialogs is challenging because of the complexity involved in rendering the dialog responsive to a vast number of varied paths through which users might desire to complete the dialog. To address this problem, we developed a toolkit for modeling and implementing task-based, mixed-initiative dialogs based on metaphors from lambda calculus. Our toolkit can automatically operationalize a dialog that involves multiple prompts and/or sub-dialogs, given a high-level dialog specification of it. Our current research entails incorporating the use of natural language to make the flexibility in communicating user utterances commensurate with that in dialog completion …

Inferring User Consumption Preferences From Social Media, Yang Li, Jing Jiang, Ting Liu

Research Collection School Of Computing and Information Systems

Social Media has already become a new arena of our lives and involved different aspects of our social presence. Users' personal information and activities on social media presumably reveal their personal interests, which offer great opportunities for many e-commerce applications. In this paper, we propose a principled latent variable model to infer user consumption preferences at the category level (e.g. inferring what categories of products a user would like to buy). Our model naturally links users' published content and following relations on microblogs with their consumption behaviors on e-commerce websites. Experimental results show our model outperforms the state-of-the-art methods significantly …

Recurrent Neural Networks With Auxiliary Labels For Cross-Domain Opinion Target Extraction, Ying Ding, Jianfei Yu, Jing Jiang

Research Collection School Of Computing and Information Systems

Opinion target extraction is a fundamental task in opinion mining. In recent years, neural network based supervised learning methods have achieved competitive performance on this task. However, as with any supervised learning method, neural network based methods for this task cannot work well when the training data comes from a different domain than the test data. On the other hand, some rule-based unsupervised methods have shown to be robust when applied to different domains. In this work, we use rule-based unsupervised methods to create auxiliary labels and use neural network models to learn a hidden representation that works well for …

Tandem 2.0: Image And Text Data Generation Application, Christopher J. Vitale

Dissertations, Theses, and Capstone Projects

First created as part of the Digital Humanities Praxis course in the spring of 2012 at the CUNY Graduate Center, Tandem explores the generation of datasets comprised of text and image data by leveraging Optical Character Recognition (OCR), Natural Language Processing (NLP) and Computer Vision (CV). This project builds upon that earlier work in a new programming framework. While other developers and digital humanities scholars have created similar tools specifically geared toward NLP (e.g. Voyant-Tools), as well as algorithms for image processing and feature extraction on the CV side, Tandem explores the process of developing a more robust and user-friendly …

The Proscriptive Principle And Logics Of Analytic Implication, Thomas M. Ferguson

Dissertations, Theses, and Capstone Projects

The analogy between inference and mereological containment goes at least back to Aristotle, whose discussion in the Prior Analytics motivates the validity of the syllogism by way of talk of parts and wholes. On this picture, the application of syllogistic is merely the analysis of concepts, a term that presupposes—through the root ἀνά + λύω —a mereological background.

In the 1930s, such considerations led William T. Parry to attempt to codify this notion of logical containment in his system of analytic implication AI. Parry’s original system AI was later expanded to the system PAI. The hallmark of Parry’s systems—and of …

Implementing Semantic Document Search Using A Bounded Random Walk In A Probabilistic Graph, Lubomir Stanchev

Computer Science and Software Engineering

Given a set of documents and an input query that is expressed using natural language, the problem of document search is retrieving all relevant documents ordered by the degree of relevance. Semantic document search fetches not only documents that contain words from the input query, but also documents that are semantically relevant. For example, the query "friendly pets" will consider documents that contain the words "dog" and "cat", among others. One way to implement semantic search is to use a probabilistic graph in which the input query is connected to the documents through paths that contain semantically similar words and …

Robot Perception Errors And Human Resolution Strategies In Situated Human-Robot Dialogue, Niels Schütte, Brian Mac Namee, John D. Kelleher

Articles

Errors in visual perception may cause problems in situated dialogues. We investigated this problem through an experiment in which human participants interacted through a natural language dialogue interface with a simulated robot.We introduced errors into the robot’s perception, and observed the resulting problems in the dialogues and their resolutions.We then introduced different methods for the user to request information about the robot’s understanding of the environment. We quantify the impact of perception errors on the dialogues, and investigate resolution attempts by users at a structural level and at the level of referring expressions.

Visualizing Incongruity: Visual Data Mining Strategies For Modeling Humor In Text, Andrew Smigaj

All Graduate Projects

The goal of this project is to investigate the use of visual data mining to model verbal humor. We explored various means of text visualization to identify key featrues of garden path jokes as compared with non jokes. With garden path jokes one interpretation is established in the setup but new information indicating some alternative interpretation triggers some resolution process leading to a new interpretation. For this project we visualize text in three novel ways, assisted by some web mining to build an informal ontology, that allow us to see the differences between garden path jokes and non jokes of …

Finding Street Gang Member Profiles On Twitter, Lakshika Balasuriya

Browse all Theses and Dissertations

The crime and violence street gangs introduce into neighborhoods is a growing epidemic in cities around the world. Today, over 1.4 million people, belonging to more than 33,000 gangs, are active in the United States, of which 88% identify themselves as being members of a street gang. With the recent popularity of social media, street gang members have established online presences coinciding with their physical occupation of neighborhoods. Recent studies report that approximately 45% of gang members participate in online offending activities such as threatening, harassing individuals, posting violent videos or attacking someone on the street for something they said …

Review Of Trends In Health Social Media Analysis, Liliya Akhtyamova, Mikhail Alexandrov, John Cardiff

Conference Papers

This paper surveys recent publications (2008-2017) on using social media data to study public health. The survey describes the main topics being discussed in forums and presents short information about methods and tools used for analysis health social media. We put especial attention on adverse drug reaction detection problem (ADR).

Adverse Drug Extraction In Twitter Data Using Convolutional Neural Network, Liliya Akhtyamova, John Cardiff, Mikhail Alexandrov

Conference Papers

The study of health-related topics on social media has become a useful tool for the early detection of the different adverse medical conditions. In particular, it concerns cases related to the treatment of mental diseases, as the effects of medications here often prove to be unpredictable. In our research, we use convolutional neural networks (CNN) with word2vec embedding to classify user comments on Twitter. The aim of the classification is to reveal adverse drug reactions of users. The results obtained are highly promising, showing the overall usefulness of neural network algorithms in this kind of tasks.

Impact Of Reviewer Social Interaction On Online Consumer Review Fraud Detection, Kunal Goswami, Younghee Park, Chungsik Song

Faculty Publications

Background Online consumer reviews have become a baseline for new consumers to try out a business or a new product. The reviews provide a quick look into the application and experience of the business/product and market it to new customers. However, some businesses or reviewers use these reviews to spread fake information about the business/product. The fake information can be used to promote a relatively average product/business or can be used to malign their competition. This activity is known as reviewer fraud or opinion spam. The paper proposes a feature set, capturing the user social interaction behavior to identify fraud. …

Gait Analysis From Wearable Devices Using Image And Signal Processing, Bradley A. Schneider

Browse all Theses and Dissertations

We present the results of analyzing gait motion in-person video taken from a commercially available wearable camera embedded in a pair of glasses. The video is analyzed with three different computer vision methods to extract motion vectors from different gait sequences from four individuals for comparison against a manually annotated ground truth dataset. Using a combination of signal processing and computer vision techniques, gait features are extracted to identify the walking pace of the individual wearing the camera and are validated using the ground truth dataset. We perform an additional data collection with both the camera and a body-worn accelerometer …

Comparison Of Two-Pass Algorithms For Dynamic Topic Modelling Based On Matrix Decompositions, John Cardiff, Gabriella Skitalinskaya, Mikhail Alexandrov

Conference Papers

In this paper we present a two-pass algorithm based on different matrix decompositions, such as LSI, PCA, ICA and NMF, which allows tracking of the evolution of topics over time. The proposed dynamic topic models as output give an easily interpreted overview of topics found in a sequentially organized set of documents that does not require further processing. Each topic is presented by a user-specified number of top-terms. Such an approach to topic modeling if applied to, for example, a news article data set, can be convenient and useful for economists, sociologists, political scientists. The proposed approach allows to achieve …

Semi-Supervised Approach To Monitoring Clinical Depressive Symptoms In Social Media, Amir Hossein Yazdavar, Hussein S. Al-Olimat, Monireh Ebrahimi, Goonmeet Bajaj, Tanvi Banerjee, Krishnaprasad Thirunarayan, Jyotishman Pathak, Amit Sheth

Computer Science and Engineering Faculty Publications

With the rise of social media, millions of people are routinely expressing their moods, feelings, and daily struggles with mental health issues on social media platforms like Twitter. Unlike traditional observational cohort studies conducted through questionnaires and self-reported surveys, we explore the reliable detection of clinical depression from tweets obtained unobtrusively. Based on the analysis of tweets crawled from users with self-reported depressive symptoms in their Twitter profiles, we demonstrate the potential for detecting clinical depression symptoms which emulate the PHQ-9 questionnaire clinicians use today. Our study uses a semi-supervised statistical model to evaluate how the duration of these symptoms …

Relatedness-Based Multi-Entity Summarization, Kalpa Gunaratna, Amir Hossein Yazdavar, Krishnaprasad Thirunarayan, Amit Sheth, Gong Cheng

Kno.e.sis Publications

Representing world knowledge in a machine processable format is important as entities and their descriptions have fueled tremendous growth in knowledge-rich information processing platforms, services, and systems. Prominent applications of knowledge graphs include search engines (e.g., Google Search and Microsoft Bing), email clients (e.g., Gmail), and intelligent personal assistants (e.g., Google Now, Amazon Echo, and Apple’s Siri). In this paper, we present an approach that can summarize facts about a collection of entities by analyzing their relatedness in preference to summarizing each entity in isolation. Specifically, we generate informative entity summaries by selecting: (i) inter-entity facts that are similar and …

The Document Similarity Network: A Novel Technique For Visualizing Relationships In Text Corpora, Dylan Baker

HMC Senior Theses

With the abundance of written information available online, it is useful to be able to automatically synthesize and extract meaningful information from text corpora. We present a unique method for visualizing relationships between documents in a text corpus. By using Latent Dirichlet Allocation to extract topics from the corpus, we create a graph whose nodes represent individual documents and whose edge weights indicate the distance between topic distributions in documents. These edge lengths are then scaled using multidimensional scaling techniques, such that more similar documents are clustered together. Applying this method to several datasets, we demonstrate that these graphs are …

Machine Learning And Natural Language Methods For Detecting Psychopathy In Textual Data, Andrew Stephen Henning

Electronic Theses and Dissertations

Among the myriad of mental conditions permeating through society, psychopathy is perhaps the most elusive to diagnose and treat. With the advent of natural language processing and machine learning, however, we have ushered in a new age of technology that provides a fresh toolkit for analyzing text and context. Because text remains the medium of choice for most personal and professional interactions, it may be possible to use textual samples from psychopaths as a means for understanding and ultimately classifying similar individuals based on the content of their language usage. This paper aims to investigate natural language processing and supervised …

Temporal Alignment Using The Incremental Unit Framework, Casey Kennington, Ting Han, David Schlangen

Computer Science Faculty Publications and Presentations

We propose a method for temporal alignments--a precondition of meaningful fusions--of multimodal systems, using the incremental unit dialogue system framework, which gives the system flexibility in how it handles alignment: either by delaying a modality for a specified amount of time, or by revoking (i.e., backtracking) processed information so multiple information sources can be processed jointly. We evaluate our approach in an offline experiment with multimodal data and find that using the incremental framework is flexible and shows promise as a solution to the problem of temporal alignment in multimodal systems.

An Alternative Approach To Training Sequence-To-Sequence Model For Machine Translation, Vivek Sah

Honors Theses

Machine translation is a widely researched topic in the field of Natural Language Processing and most recently, neural network models have been shown to be very effective at this task. The model, called sequence-to-sequence model, learns to map an input sequence in one language to a vector of fixed dimensionality and then map that vector to an output sequence in another language without any human intervention provided that there is enough training data. Focusing on English-French translation, in this paper, I present a way to simplify the learning process by replacing English input sentences by word-by-word translation of those sentences. …

Smart Image Search System Using Personalized Semantic Search Method, Fangyu Zhang

Electronic Theses and Dissertations

Due to the emerge in huge numbers of information on the internet nowadays, search technologies are widely used in various fields. Achieving the most relevant search result for the users becomes a big challenge now. While the traditional semantic search technologies seem to achieve the most relevant search result, however, it faces two problems: one is the one-size-fits-all problem, and another is low efficiency. The purpose of this research is to build a Smart Image Search System by using the personalized semantic search method to solve those problems. The personalized semantic search method makes the search system avoids the one-size-fits-all …