UMD Theses and Dissertations

Permanent URI for this collectionhttp://hdl.handle.net/1903/3

New submissions to the thesis/dissertation collections are added automatically as they are received from the Graduate School. Currently, the Graduate School deposits all theses and dissertations from a given semester after the official graduation date. This means that there may be up to a 4 month delay in the appearance of a given thesis/dissertation in DRUM.

More information is available at Theses and Dissertations at University of Maryland Libraries.

Browse

Filters

2019 - 201912020 - 20231

Settings

Subject: search.filters.subject.context-aware

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Towards Multimodal and Context-Aware Emotion Perception
    (2023) Mittal, Trisha; Manocha, Dinesh Dr.; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    Human emotion perception is a part of affective computing, a branch of computing that studies and develops systems and devices that can recognize, interpret, process, and simulate human affects. Research in human emotion perception, however, has been mostly restricted to psychology-based literature which explores the theoretical aspects of emotion perception, but does not touch upon its practical applications. For instance, human emotion perception plays a pivotal role in an extensive array of sophisticated intelligent systems, encompassing domains such as behavior prediction, social robotics, medicine, surveillance, and entertainment. In order to deploy emotion perception in these applications, extensive research in psychology has demonstrated that humans not only perceive emotions and behavior through diverse human modalities but also glean insights from situational and contextual cues. This dissertation not only enhances the capabilities of existing human emotion perception systems but also forges novel connections between emotion perception and multimedia analysis, social media analysis, and multimedia forensics. Specifically, this work introduces two innovative algorithms that revolutionize the construction of human emotion perception models. These algorithms are then applied to detect falsified multimedia, understand human behavior and psychology on social media networks, and extract the intricate array of emotions evoked by movies. In the first part of this dissertation, we delve into two unique approaches to advance emotion perception models. The first approach capitalizes on the power of multiple modalities to perceive human emotion. The second approach leverages the contextual information, such as the background scene, diverse modalities of the human subject, and intricate socio-dynamic inter-agent interactions. These elements converge to predict perceived emotions with better accuracy, culminating in the development of context-aware human emotion perception models. In the second part of this thesis, we forge connections between emotion perception and three prominent domains of artificial intelligence applications. These domains include video manipulations and deepfake detection, multimedia content analysis, and user behavior analysis on social media platforms. Drawing inspiration from emotion perception, we conceptualize enriched solutions that push the conventional boundaries and redefine the possibilities within these domains. All experiments in this dissertation have been conducted on all state-of-the-art emotion perception datasets, including IEMOCAP, CMU-MOSEI, EMOTIC, SENDv1, MovieGraphs, LIRIS-ACCEDE, DF-TIMIT, DFDC, Intentonomy, MDID, and MET-Meme. In fact, we propose three additional datasets to this list, namely GroupWalk, VideoSham and IntentGram. In addition to providing quantitative results to validate our claims, we conduct user evaluations where applicable, serving as a compelling testament to the remarkable outcomes of our experiments.
  • Thumbnail Image
    Item
    STRUCTANT: A CONTEXT-AWARE TASK MANAGEMENT FRAMEWORK FOR HETEROGENEOUS COMPUTATIONAL ENVIRONMENTS
    (2019) Pachulski, Andrew J; Agrawala, Ashok; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    The Internet of Things has produced a plethora of devices, systems, and networks able to produce, transmit, and process data at unprecedented rates. These data can have tremendous value for businesses, organizations, and researchers who wish to better serve an audience or understand a topic. Pipelining is a common technique used to automate the scraping, processing, transport, and analytic steps necessary for collecting and utilizing these data.Each step in a pipeline may have specific physical, virtual, and organizational processing requirements that dictate when the step can run and what machines can run it. Physical processing requirements may include hardware specific computing capabilities such as the presence of Graphics Processing Units (GPU), memory capacity, and specific CPU instruction sets. Virtual processing requirements may include job precedence, machine architecture, availability of input datasets, runtime libraries, and executable code. Organizational processing requirements may include encryption standards for data transport and data at rest, physical server security, and monetary budget constraints. Moreover, these processing requirements may have dynamic or temporal properties not known until schedule time.These processing requirements can greatly impact the ability organizations to use these data. Despite the popularity of Big Data and cloud computing and the plethora of tools they provide, organizations still face challenges when attempting to adopt these solutions. These challenges include the need to recreate the pipeline, cryptic configuration parameters, and inability to support rapid deployment and modification for data exploration. Prior work has focused on solutions that apply only to specific steps, platforms, or algorithms in the pipeline, without considering the abundance of information that describes the processing environment and operations.In this dissertation, we present Structant, a context-aware task management framework and scheduler that helps users manage complex physical, virtual, and organizational processing requirements. Structant models jobs, machines, links, and datasets by storing contextual information for each entity in the Computational Environment. Through inference of this contextual information, Structant creates mappings of jobs to resources that satisfy all relevant processing requirements. As jobs execute, Structant observes performance and creates runtime estimates for new jobs based on prior execution traces and relevant context selection. Using runtime estimates, Structant can schedule jobs with respect to dynamic and temporal processing requirements.We present results from three experiments to demonstrate how Structant can aid a user in running both simple and complex pipelines. In our first experiment, we demonstrate how Structant can schedule data collection, processing, and movement with virtual processing requirements to facilitate forward prediction of communities at risk for opioid epidemics. In our second experiment, we demonstrate how Structant can profile operations and obey temporal organizational policies to schedule data movement with fewer preemptions than two naive scheduling algorithms. In our third experiment, we demonstrate how Structant can acquire external contextual information from server room monitors and maintain regulatory compliance of the processing environment by shutting down machines according to a predetermined pipeline.