Social media has become a primary platform for real-time information sharing among users. News on social media spreads faster than traditional outlets and millions of users turn to this platform to receive the latest updates on major events especially disasters. Social media bridges the gap between the people who are affected by disasters, volunteers who offer contributions, and first responders. On the other hand, social media is a fertile ground for malicious users who purposefully disturb the relief processes facilitated on social media. These malicious users take advantage of social bots to overrun social media posts with fake images, rumors, and false information. This process causes distress and prevents actionable information from reaching the affected people. Social bots are automated accounts that are controlled by a malicious user and these bots have become prevalent on social media in recent years.

In spite of existing efforts towards understanding and removing bots on social media, there are at least two drawbacks associated with the current bot detection algorithms: general-purpose bot detection methods are designed to be conservative and not label a user as a bot unless the algorithm is highly confident and they overlook the effect of users who are manipulated by bots and (unintentionally) spread their content. This study is trifold. First, I design a Machine Learning model that uses content and context of social media posts to detect actionable ones among them; it specifically focuses on tweets in which people ask for help after major disasters. Second, I focus on bots who can be a facilitator of malicious content spreading during disasters. I propose two methods for detecting bots on social media with a focus on the recall of the detection. Third, I study the characteristics of users who spread the content of malicious actors. These features have the potential to improve methods that detect malicious content such as fake news.

Graph is a ubiquitous data structure, which appears in a broad range of real-world scenarios. Accordingly, there has been a surge of research to represent and learn from graphs in order to accomplish various machine learning and graph analysis tasks. However, most of these efforts only utilize the graph structure while nodes in real-world graphs usually come with a rich set of attributes. Typical examples of such nodes and their attributes are users and their profiles in social networks, scientific articles and their content in citation networks, protein molecules and their gene sets in biological networks as well as web pages and their content on the Web. Utilizing node features in such graphs---attributed graphs---can alleviate the graph sparsity problem and help explain various phenomena (e.g., the motives behind the formation of communities in social networks). Therefore, further study of attributed graphs is required to take full advantage of node attributes.

In the wild, attributed graphs are usually unlabeled. Moreover, annotating data is an expensive and time-consuming process, which suffers from many limitations such as annotators’ subjectivity, reproducibility, and consistency. The challenges of data annotation and the growing increase of unlabeled attributed graphs in various real-world applications significantly demand unsupervised learning for attributed graphs.

In this dissertation, I propose a set of novel models to learn from attributed graphs in an unsupervised manner. To better understand and represent nodes and communities in attributed graphs, I present different models in node and community levels. In node level, I utilize node features as well as the graph structure in attributed graphs to learn distributed representations of nodes, which can be useful in a variety of downstream machine learning applications. In community level, with a focus on social media, I take advantage of both node attributes and the graph structure to discover not only communities but also their sentiment-driven profiles and inter-community relations (i.e., alliance, antagonism, or no relation). The discovered community profiles and relations help to better understand the structure and dynamics of social media.

Millions of users leave digital traces of their political engagements on social media platforms every day. Users form networks of interactions, produce textual content, like and share each others' content. This creates an invaluable opportunity to better understand the political engagements of internet users. In this proposal, I present three algorithmic solutions to three facets of online political networks; namely, detection of communities, antagonisms and the impact of certain types of accounts on political polarization. First, I develop a multi-view community detection algorithm to find politically pure communities. I find that word usage among other content types (i.e. hashtags, URLs) complement user interactions the best in accurately detecting communities.

Second, I focus on detecting negative linkages between politically motivated social media users. Major social media platforms do not facilitate their users with built-in negative interaction options. However, many political network analysis tasks rely on not only positive but also negative linkages. Here, I present the SocLSFact framework to detect negative linkages among social media users. It utilizes three pieces of information; sentiment cues of textual interactions, positive interactions, and socially balanced triads. I evaluate the contribution of each three aspects in negative link detection performance on multiple tasks.

Third, I propose an experimental setup that quantifies the polarization impact of automated accounts on Twitter retweet networks. I focus on a dataset of tragic Parkland shooting event and its aftermath. I show that when automated accounts are removed from the retweet network the network polarization decrease significantly, while a same number of accounts to the automated accounts are removed randomly the difference is not significant. I also find that prominent predictors of engagement of automatically generated content is not very different than what previous studies point out in general engaging content on social media. Last but not least, I identify accounts which self-disclose their automated nature in their profile by using expressions such as bot, chat-bot, or robot. I find that human engagement to self-disclosing accounts compared to non-disclosing automated accounts is much smaller. This observational finding can motivate further efforts into automated account detection research to prevent their unintended impact.

Parents fulfill a pivotal role in early childhood development of social and communication

skills. In children with autism, the development of these skills can be delayed. Applied

behavioral analysis (ABA) techniques have been created to aid in skill acquisition.

Among these, pivotal response treatment (PRT) has been empirically shown to foster

improvements. Research into PRT implementation has also shown that parents can be

trained to be effective interventionists for their children. The current difficulty in PRT

training is how to disseminate training to parents who need it, and how to support and

motivate practitioners after training.

Evaluation of the parents’ fidelity to implementation is often undertaken using video

probes that depict the dyadic interaction occurring between the parent and the child during

PRT sessions. These videos are time consuming for clinicians to process, and often result

in only minimal feedback for the parents. Current trends in technology could be utilized to

alleviate the manual cost of extracting data from the videos, affording greater

opportunities for providing clinician created feedback as well as automated assessments.

The naturalistic context of the video probes along with the dependence on ubiquitous

recording devices creates a difficult scenario for classification tasks. The domain of the

PRT video probes can be expected to have high levels of both aleatory and epistemic

uncertainty. Addressing these challenges requires examination of the multimodal data

along with implementation and evaluation of classification algorithms. This is explored

through the use of a new dataset of PRT videos.

The relationship between the parent and the clinician is important. The clinician can

provide support and help build self-efficacy in addition to providing knowledge and

modeling of treatment procedures. Facilitating this relationship along with automated

feedback not only provides the opportunity to present expert feedback to the parent, but

also allows the clinician to aid in personalizing the classification models. By utilizing a

human-in-the-loop framework, clinicians can aid in addressing the uncertainty in the

classification models by providing additional labeled samples. This will allow the system

to improve classification and provides a person-centered approach to extracting

multimodal data from PRT video probes.

Causality analysis is the process of identifying cause-effect relationships among variables. This process is challenging because causal relationships cannot be tested solely based on statistical indicators as additional information is always needed to reduce the ambiguity caused by factors beyond those covered by the statistical test. Traditionally, controlled experiments are carried out to identify causal relationships, but recently there is a growing interest in causality analysis with observational data due to the increasing availability of data and tools. This type of analysis will often involve automatic algorithms that extract causal relations from large amounts of data and rely on expert judgment to scrutinize and verify the relations. Over-reliance on these automatic algorithms is dangerous because models trained on observational data are susceptible to bias that can be difficult to spot even with expert oversight. Visualization has proven to be effective at bridging the gap between human experts and statistical models by enabling an interactive exploration and manipulation of the data and models. This thesis develops a visual analytics framework to support the interaction between human experts and automatic models in causality analysis. Three case studies were conducted to demonstrate the application of the visual analytics framework in which feature engineering, insight generation, correlation analysis, and causality inspections were showcased.

Time series forecasting is the prediction of future data after analyzing the past data for temporal trends. This work investigates two fields of time series forecasting in the form of Stock Data Prediction and the Opioid Incident Prediction. In this thesis, the Stock Data Prediction Problem investigates methods which could predict the trends in the NYSE and NASDAQ stock markets for ten different companies, nine of which are part of the Dow Jones Industrial Average (DJIA). A novel deep learning model which uses a Generative Adversarial Network (GAN) is used to predict future data and the results are compared with the existing regression techniques like Linear, Huber, and Ridge regression and neural network models such as Long-Short Term Memory (LSTMs) models.

In this thesis, the Opioid Incident Prediction Problem investigates methods which could predict the location of future opioid overdose incidences using the past opioid overdose incidences data. A similar deep learning model is used to predict the location of the future overdose incidences given the two datasets of the past incidences (Connecticut and Cincinnati Opioid incidence datasets) and compared with the existing neural network models such as Convolution LSTMs, Attention-based Convolution LSTMs, and Encoder-Decoder frameworks. Experimental results on the above-mentioned datasets for both the problems show the superiority of the proposed architectures over the standard statistical models.

With the development of modern technological infrastructures, such as social networks or the Internet of Things (IoT), data is being generated at a speed that is never before seen. Analyzing the content of this data helps us further understand underlying patterns and discover relationships among different subsets of data, enabling intelligent decision making. In this thesis, I first introduce the Low-rank, Win-dowed, Incremental Singular Value Decomposition (SVD) framework to inclemently maintain SVD factors over streaming data. Then, I present the Group Incremental Non-Negative Matrix Factorization framework to leverage redundancies in the data to speed up incremental processing. They primarily tackle the challenges of using factorization models in the scenarios with streaming textual data. In order to tackle the challenges in improving the effectiveness and efficiency of generative models in this streaming environment, I introduce the Incremental Dynamic Multiscale Topic Model framework, which identifies multi-scale patterns and their evolutions within streaming datasets. While the latent factor models assume the linear independence in the latent factors, the generative models assume the observation is generated from a set of latent variables with various distributions. Furthermore, some models may not be accessible or their underlying structures are too complex to understand, such as simulation ensembles, where there may be thousands of parameters with a huge parameter space, the only way to learn information from it is to execute real simulations. When performing knowledge discovery and decision making through data- and model-driven simulation ensembles, it is expensive to operate these ensembles continuously at large scale, due to the high computational. Consequently, given a relatively small simulation budget, it is desirable to identify a sparse ensemble that includes the most informative simulations, while still permitting effective exploration of the input parameter space. Therefore, I present Complexity-Guided Parameter Space Sampling framework, which is an intelligent, top-down sampling scheme to select the most salient simulation parameters to execute, given a limited computational budget. Moreover, I also present a Pivot-Guided Parameter Space Sampling framework, which incrementally maintains a diverse ensemble of models of the simulation ensemble space and uses a pivot guided mechanism for future sample selection.

Fraud is defined as the utilization of deception for illegal gain by hiding the true nature of the activity. While organizations lose around $3.7 trillion in revenue due to financial crimes and fraud worldwide, they can affect all levels of society significantly. In this dissertation, I focus on credit card fraud in online transactions. Every online transaction comes with a fraud risk and it is the merchant's liability to detect and stop fraudulent transactions. Merchants utilize various mechanisms to prevent and manage fraud such as automated fraud detection systems and manual transaction reviews by expert fraud analysts. Many proposed solutions mostly focus on fraud detection accuracy and ignore financial considerations. Also, the highly effective manual review process is overlooked. First, I propose Profit Optimizing Neural Risk Manager (PONRM), a selective classifier that (a) constitutes optimal collaboration between machine learning models and human expertise under industrial constraints, (b) is cost and profit sensitive. I suggest directions on how to characterize fraudulent behavior and assess the risk of a transaction. I show that my framework outperforms cost-sensitive and cost-insensitive baselines on three real-world merchant datasets. While PONRM is able to work with many supervised learners and obtain convincing results, utilizing probability outputs directly from the trained model itself can pose problems, especially in deep learning as softmax output is not a true uncertainty measure. This phenomenon, and the wide and rapid adoption of deep learning by practitioners brought unintended consequences in many situations such as in the infamous case of Google Photos' racist image recognition algorithm; thus, necessitated the utilization of the quantified uncertainty for each prediction. There have been recent efforts towards quantifying uncertainty in conventional deep learning methods (e.g., dropout as Bayesian approximation); however, their optimal use in decision making is often overlooked and understudied. Thus, I present a mixed-integer programming framework for selective classification called MIPSC, that investigates and combines model uncertainty and predictive mean to identify optimal classification and rejection regions. I also extend this framework to cost-sensitive settings (MIPCSC) and focus on the critical real-world problem, online fraud management and show that my approach outperforms industry standard methods significantly for online fraud management in real-world settings.

Networks naturally appear in many high-impact applications. The simplest model of networks is single-layered networks, where the nodes are from the same domain and the links are of the same type. However, as the world is highly coupled, nodes from different application domains tend to be interdependent on each other, forming a more complex network model called multi-layered networks.

Among the various aspects of network studies, network connectivity plays an important role in a myriad of applications. The diversified application areas have spurred numerous connectivity measures, each designed for some specific tasks. Although effective in their own fields, none of the connectivity measures is generally applicable to all the tasks. Moreover, existing connectivity measures are predominantly based on single-layered networks, with few attempts made on multi-layered networks.

Most connectivity analyzing methods assume that the input network is static and accurate, which is not realistic in many applications. As real-world networks are evolving, their connectivity scores would vary by time as well, making it imperative to keep track of those changing parameters in a timely manner. Furthermore, as the observed links in the input network may be inaccurate due to noise and incomplete data sources, it is crucial to infer a more accurate network structure to better approximate its connectivity scores.

The ultimate goal of connectivity studies is to optimize the connectivity scores via manipulating the network structures. For most complex measures, the hardness of the optimization problem still remains unknown. Meanwhile, current optimization methods are mainly ad-hoc solutions for specific types of connectivity measures on single-layered networks. No optimization framework has ever been proposed to tackle a wider range of connectivity measures on complex networks.

In this thesis, an in-depth study of connectivity measures, inference, and optimization problems will be proposed. Specifically, a unified connectivity measure model will be introduced to unveil the commonality among existing connectivity measures. For the connectivity inference aspect, an effective network inference method and connectivity tracking framework will be described. Last, a generalized optimization framework will be built to address the connectivity minimization/maximization problems on both single-layered and multi-layered networks.

In the artificial intelligence literature, three forms of reasoning are commonly employed to understand agent behavior: inductive, deductive, and abductive.  More recently, data-driven approaches leveraging ideas such as machine learning, data mining, and social network analysis have gained popularity. While data-driven variants of the aforementioned forms of reasoning have been applied separately, there is little work on how data-driven approaches across all three forms relate and lend themselves to practical applications. Given an agent behavior and the percept sequence, how one can identify a specific outcome such as the likeliest explanation? To address real-world problems, it is vital to understand the different types of reasonings which can lead to better data-driven inference.  

This dissertation has laid the groundwork for studying these relationships and applying them to three real-world problems. In criminal modeling, inductive and deductive reasonings are applied to early prediction of violent criminal gang members. To address this problem the features derived from the co-arrestee social network as well as geographical and temporal features are leveraged. Then, a data-driven variant of geospatial abductive inference is studied in missing person problem to locate the missing person. Finally, induction and abduction reasonings are studied for identifying pathogenic accounts of a cascade in social networks.