A5067722075- Data Visualization and Analytics
- Topic Modeling
- Advanced Text Analysis Techniques
- Video Analysis and Summarization
- Natural Language Processing Techniques
- Context-Aware Activity Recognition Systems
- Multimodal Machine Learning Applications
- Image Retrieval and Classification Techniques
- Energy Efficient Wireless Sensor Networks
- Geotechnical Engineering and Soil Mechanics
- Advanced Image and Video Retrieval Techniques
- Multimedia Communication and Technology
- Geotechnical Engineering and Soil Stabilization
- Geotechnical Engineering and Underground Structures
- Speech and dialogue systems
- Complex Network Analysis Techniques
- Semantic Web and Ontologies
- IoT-based Smart Home Systems
- IoT and Edge/Fog Computing
- Speech Recognition and Synthesis
- Geotechnical and construction materials studies
- Music and Audio Processing
- Modular Robots and Swarm Intelligence
- Time Series Analysis and Forecasting
- Anomaly Detection Techniques and Applications
York University
2019-2024
East Stroudsburg University
2023
University of Virginia
2009-2019
Stanford University
2017-2018
University of British Columbia
2014-2017
University of Southern California
2017
Drexel University
2017
Tallinn University of Technology
2017
Bauhaus-Universität Weimar
2017
Technical University of Darmstadt
2017
Interactive visual data analysis is most productive when users can focus on answering the questions they have about their data, rather than focusing how to operate interface tool. One viable approach engaging in interactive conversations with a natural language visualizations. These interfaces potential be both more expressive and accessible other interaction paradigms. We explore principles from pragmatics applied flow of analytical conversations, using as an input modality. evaluate...
People often use charts to analyze data, answer questions and explain their answers others. In a formative study, we find that such human-generated explanations commonly refer visual features of charts. Based on this developed an automatic chart question answering pipeline generates describing how the was obtained. Our first extracts data encodings from input Vega-Lite chart. Then, given natural language about chart, it transforms references attributes into data. It next applies...
Abstract The Query-Focused Text Summarization (QFTS) task aims at building systems that generate the summary of text document(s) based on given query. A key challenge in addressing this is lack large labeled data for training summarization model. In article, we address by exploring a series domain adaptation techniques. Given recent success pre-trained transformer models wide range natural language processing tasks, utilize such to abstractive summaries QFTS both single-document and...
Sleep monitoring is very important for elderly people as inadequate and irregular sleep are often related to serious diseases such depression diabetes. In many cases, it necessary monitor the body positions movements made while sleeping because of their relationships particular (i.e., apnea restless legs syndrome). Analyzing during also helps in determining quality patterns. This paper presents a system based on WISP platform - active RFID-based sensors equipped with accelerometers. We show...
As wireless sensor networks mature, they are increasingly being used in real-time applications. Many of these applications require reliable transmission within latency bounds. Achieving this goal is very difficult because link burstiness and interference. Based on significant empirical evidence 21 days over 3,600,000 packets per link, we propose a scheduling algorithm that produces bounds the periodic streams accounts for both bursts The solution achieved through definition new metric Bmax...
We present AALO: a novel Activity recognition system for single person smart homes using Active Learning in the presence of Overlapped activities. AALO applies data mining techniques to cluster in-home sensor firings so that each represents instances same activity. Users only need
Abstract Today it is quite common for people to exchange hundreds of comments in online conversations (e.g., blogs). Often, can be very difficult analyze and gain insights from such long conversations. To address this problem, we present a visual text analytic system that tightly integrates interactive visualization with novel mining summarization techniques fulfill information needs users exploring At first, perform user requirement analysis the domain blog derive set design principles....
There is increasing evidence that the effectiveness of information visualization techniques can be impacted by particular needs and abilities each user. This suggests it important to investigate systems dynamically adapt In this paper, we address question how adapt. particular, present a study evaluate variety visual prompts, called "interventions", performed on help users process it. Our results show some tested interventions perform better than condition in which no intervention provided,...
In the last decade, there has been an exponential growth of asynchronous online conversations thanks to rise social media. Analyzing and gaining insights from such can be quite challenging for a user, especially when discussion becomes very long. A promising solution this problem is topic modeling, since it may help user quickly understand what was discussed in long conversation explore comments interest. However, results modeling noisy not match user's current information needs. To address...
Information visualizations such as bar charts and line are very popular for exploring data communicating insights. Interpreting making sense of can be challenging some people, those who visually impaired or have low visualization literacy. In this work, we introduce a new dataset present neural model automatically generating natural language summaries charts. The generated provide an interpretation the chart convey key insights found within that chart. Our is developed by extending...
Abstract There is increasing evidence that user characteristics can have a significant impact on visualization effectiveness, suggesting visualizations could be designed to better fit each user's specific needs. Most studies date, however, looked at static visualizations. Studies considering interactive only limited number of characteristics, and consider either low‐level tasks (e.g., value retrieval), or high‐level (in particular: discovery), but not both. This paper contributes this line...
Advances in wireless sensor networks have enabled the monitoring of daily activities elderly people. The goal these applications is to learn normal behavior terms and look for any deviation, i.e., Anomalies, so that alerts can be sent relatives or caregivers. However, human very complex, many existing anomaly detection systems are too simplistic which cause false alarms, resulting unreliable systems. We present Holmes, a comprehensive system in-home activities. Holmes accurately learns...
Document authors commonly use tables to support arguments presented in the text. But, because are usually separate from main body text, readers must split their attention between different parts of document. We present an interactive document reader that automatically links text with corresponding table cells. Readers can select a sentence (or cells) and our highlights relevant cells sentences). provide automatic pipeline for extracting such references existing PDF documents combines...
We present a search engine for D3 visualizations that allows queries based on their visual style and underlying structure. To build the we crawl collection of 7860 from Web deconstruct each one to recover its data, data-encoding marks encodings describing how data is mapped attributes marks. also extract axes other non-data-encoding (e.g., typeface, background color). Our indexes this structure information as well metadata about webpage containing chart. show visualization developers can...
Web-based data visualizations have become very popular for exploring and communicating insights. Newspapers, journals, reports regularly publish to tell compelling stories with data. Unfortunately, most are inaccessible readers visual impairments. For many charts on the web, there no accompanying alternative (alt) texts, even if such texts exist they do not adequately describe important insights from charts. To address problem, we first interviewed 15 blind users understand their challenges...
The modelling of non-linear stifaffess, based on elastic characteristic moduli for stiff geomaterials, is proposed, the basis results a large number triaxial compression tests sedimentary soft rocks, tuff, weathered granite, cement-treated soils, densely compacted gravels and sands. Axial strains from less than 0·001% to about 1% were measured by using local gauge, while Young's E sbp e obtained applying series small unload/reload cycles axial stress during each monotonic loading test....
This paper describes a novel group-based programming abstraction called "Bundle" for cyber-physical systems (CPS). Similar to other abstractions, Bundle creates logical collections of sensing devices. However, previous abstractions were focused on wireless sensor networks (WSNs) and did not address key aspects CPS. Bundles elevate the domain from single WSN complex by allowing applications involving multiple CPSs that are controlled different administrative domains support mobility both...
Aligning visual features with language embeddings is a key challenge in vision-language models (VLMs). The performance of such hinges on having good connector that maps generated by vision encoder to shared embedding space the LLM while preserving semantic similarity. Existing connectors, as multilayer perceptrons (MLPs), often produce out-of-distribution or noisy inputs, leading misalignment between modalities. In this work, we propose novel vision-text alignment method, AlignVLM, weighted...
A comprehensive series of triaxial tests were performed on granular materials to study the stress-dependence characteristics small-strain stiffness (i.e. quasi-elastic Young's modulus). The large prismatic specimens measuring both axial and lateral strains locally. at a given stress state was measured from stress–strain response obtained by applying very small-amplitude cyclic normal stresses, in which single-amplitude major principal strain less than 0·002%. specimen varied along various...
Abstract An advanced automated triaxial testing system developed to characterize elastic behavior of soils is described. Small axial and lateral strains a large rectangular prismatic specimen were measured locally at various stress states along specified path, which was achieved by automatically controlling load confining pressure. Elastic deformation applying very small cycles or each given neutral state. Typical test data for air-dried sand are presented. It observed that the Young's...