Graduation Year

2022

Document Type

Dissertation

Degree

Ph.D.

Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Electrical Engineering

Major Professor

Ravi Sankar, Ph.D.

Committee Member

Nasir Ghani, Ph.D.

Committee Member

Ismail Uysal, Ph.D.

Committee Member

Manish Agrawal, Ph.D.

Committee Member

Inho Ra, Ph.D.

Keywords

Binary Search Tree, Cooperation Strategy, Influence Network, Optimization, Machine Learning, Graph Network, Social Network

Abstract

Social networks have attracted increasing attention from both physical and social scientists. Social networks are essential elements in societies, serving as channels for exchanging various benefits, such as innovation, information, and social support. Moreover, research in social networks helps explain macro-level social phenomena, such as social polarization and social contagion. An understanding of social networks has significant implications, such as improving social welfare and political participation. Modeling social network formation has typically employed game theory or agent-based modeling. These studies typically propose simple and tractable micro-level rules for link formation mechanisms and show that these rules have implications for known macro-level properties. Statistics and econometrics have also used game theory to model empirical networks, but they typically have been focused on estimating and identifying the effects of interest, such as racial segregation. To date, these models have not been capable of accounting for the effects of broad heterogeneity among individuals; therefore, they lack predictive power for link formation in complex, real-world networks. This divergence is filled by cooperative techniques by applying game theory and casual inference techniques on severe weather prediction and disease spread in our work with consideration of heterogeneity, predictability of link formation and node characteristics.

The recent trend of dependence on the social network for information abstraction and propagation has a cumulative effect on critical response. The content and reliability of data are substantiated by acquiring data from a network of Twitter users. It captures the engaged multiple user behavior to formulate and diffuse the connected information across the channel. The objective is to identify a bridge between different data sources for event anomalies. This dissertation proposes a novel approach towards identifying the sublevel anomalies and predictive investigation towards the use of Twitter’s social data in the extreme weather scenario and disease spread. We performed qualitative analyses by gathering data from social media and weather data websites and government websites. We also focused on a casual cooperation model outlined from social data with the help of survey data. The cooperation model encompasses cooperative attention to detect possible anomaly in an event. Various analysis methods are proposed to aggregate diffused information from the social network to generate influence data. This research also proposes the determination of spread through cooperative learning with the help of disease spread model. The analyses result further identify connected user acknowledgment for dominant information in the public domain. This information is mapped by applying a convolutional neural network for a physical sensor dataset to detect weather anomalies. Moreover, we exploited the causal inference technique to determine smart policy on influence data. The results show that our proposed method can predict critical events with high precision at the accuracy of 67% during extreme weather emergency scenarios specifically studied on hurricane IRMA.

Cooperative attention outlines the new paradigm for finding the cause of epidemic disease spread. It can be derived from the social data with the help of survey data. The cooperative attention enables the rationale to detect possible anomaly in an event by formulating the spread variable to determine disease spread rate decision score. This research proposes the determination of spread through cooperative learning with the help of disease spread model. We used game theory to define cooperative strategy and analyzed the determined dynamic states with the help of control algorithm. This model is a four-stage model to determine rewards by identifying the semantic cooperation with spread model to identify events, infection factor, location spread, and change in spread rate. Our model proposes new approach to define data cooperation by finding dynamic variable of spread and optimal cooperative strategy for the analysis of COVID 19 pandemic spread across Unites States. Our analysis successfully identified the spread rate of disease from social data with an accuracy of 81% and can dynamically optimize the decision model with O(n2).

The research also presents the development of systems for improved source selection in a process that creates real time categorization of events using only posts collected through various sensing applications that use social networks (such as Twitter or other mass dissemination networks) for reporting. The system recognizes critical instances in applications and simply views essential information from users (either by explicit user action or by default, as on Twitter) within the event and provides a textual description. As a result, social networks open unprecedented possibilities for creating sensing applications by representing a set of tweets generated in a limited timeframe as a weighted network for influence concerning users. Obtaining data from a network of social site users substantiates the quality and dependability of data. It collects many users' dynamic behavior to construct and disseminate related information across the channel. The goal is to find a link between various data sources for event abnormalities. By detecting sublevel anomalies using a convex optimization framework; the system recognizes rapid changes in the graphs' nodes and edge weights to pinpoint anomalies inside an event. This research investigates the merits of diversified data sources and developed graphical relations of information learning by correlation between social and different data sources by understanding the heterogeneity and homophily in a network with optimal accuracy.

Download

Share

COinS