Social media provides abundant, naturally occurring data on communications among people that can reveal their behavior. Recent technological advancements have made possible the rapid processing of social media data with the potential to learn more about human behavior. In order to do so, however, there is a need for a comprehensive and integrative theoretical model of human behavior as expressed in social media. In addition, such a model should provide specific avenues for staging behavioral interventions. The objective of this paper is to address this deficiency in the context of behaviors that occur in response to significant events. Relevant theories of human behavior are presented as well as a methodology, utilizing natural language processing and social network analysis, for measuring elements of human behavior. The methodology provides numerical measures for each element, which serve as input data for the multivariate statistical model used to answer research questions in a case study of human behavior in response to a natural disaster. For this case study, three behaviors associated with the warning response process were modeled: 1) obtain and propagate the warning; 2) seek additional information/confirmation; and 3) take the prescribed action. Emergency managers can use the findings of the research to help ensure that people obtain critical information and take prescribed actions, such as evacuating a flood zone. Once the dynamics of human behavior as expressed in social media are understood, models can be built to predict human behaviors well in advance of their recommended or anticipated occurrence.

Information and communication technology (ICT) has increasingly important implications for our everyday lives, with the potential to both solve existing social problems and create new ones. This article focuses on one particular group of ICT professionals, computational modelers, and explores how these ICT professionals perceive their own societal responsibilities. Specifically, the article uses a mixed‐method approach to look at the role of professional codes of ethics and explores the relationship between modelers’ experiences with, and attitudes toward, codes of ethics and their values. Statistical analysis of survey data reveals a relationship between modelers’ values and their attitudes and experiences related to codes of ethics. Thematic analysis of interviews with a subset of survey participants identifies two key themes: that modelers should be faithful to the reality and values of users and that codes of ethics should be built from the bottom up. One important implication of the research is that those who value universalism and benevolence may have a particular duty to act on their values and advocate for, and work to develop, a code of ethics.

We study the possibility of predicting a social protest (planned, or unplanned) based on social media messaging. We consider the process called mobilization, described in the literature as the precursor of participation. Mobilization includes four stages: being sympathetic to the cause, being aware of the movement, motivation to take part and ability to participate. We suggest that expressions of mobilization in communications of individuals may be used to predict the approaching protest. We have utilized several Natural Language Processing techniques to create a methodology to identify mobilization in social media communication. Results of experimentation with Twitter data collected before and during the 2015 Baltimore events and the information on actual protests taken from news media show a correlation over time between volume of Twitter communications related to mobilization and occurrences of protest at certain geographical locations. We conclude with discussion of possible theoretical explanations and practical applications of these results.

Dian Yu, Yulia Tyshchuk, Heng Ji, William Wallace. Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2015.

We study the relationship between the level of chatter on a social medium (like Twitter) and the level of the observed actions related to the chatter. For example, in a disaster, how does relief-donation chatter on Twitter correlate with the dollar amount received? One hypothesis is that a fraction of those who act will also tweet about it, which implies linear scaling, action ∝ chatter. On the other hand, if there is a contagion effect (those who tweet about donation incite others to donate) and these incited donors tend to be "quiet" and not broadcast their actions, then we expect superlinear scaling,

Leadership has traditionally been viewed as a relationship between the leaders and their subordinates. Recent research has shown the significance of the social aspects of leadership. This view forms the basis for the research reported on in this paper. Social network analysis provides socio-metric measures of leadership, which can be associated with traditional measures of leadership. The research evaluates U.S. Military cadets' social networks of friendship, trust, and reputed leadership in concert with leadership attributes, organizational identity, and attitude. The significance of reputed leaders' social position is ascertained. Additionally, the channels and extent of influence of reputed d leaders over other cadets are assessed. A significant result was that leaders are not spreading their influence over the friendship network but over the leadership reputation network. Leaders have a particular influence on those who perceive them as leaders when it comes to leadership style and performance. This result contributes to our knowledge of “social leadership”, in particular, the ability to manage the social dynamics inherent in leadership.

Twitter is presently utilized as a channel of communication and information dissemination. At present, government and non-government emergency management organizations utilize Twitter to disseminate emergency relevant information. However, these organizations have limited ability to evaluate the Twitter communication in order to discover communication patterns, key players, and messages that are being propagated through Twitter regarding the event. More importantly there is a general lack of knowledge of who are the individuals or organizations that disseminate warning information, provide confirmations of an event and associated actions, and urge others to take action. This paper presents a methodology that shows how Natural Language Processing (NLP) and Social Network Analysis (SNA) can aid in addressing these issues. The methodology, in addition to qualitative data collected during on-site interviews and publicly available information, was successfully applied to a Twitter data set collected during 2011 Japan tsunami. NLP techniques were applied to extract actionable messages. Based on the messages extracted by NLP, SNA was used to construct a network of actionable messages. While SNA discovered communities and extracted the community leaders, NLP was used to determine the behavior of the community members and the role of the community leaders. Therefore, the proposed methodology automatically finds communities, evaluates its members' behaviors, and authenticates cohesive behaviors of the community members during emergencies. Moreover, the methodology efficiently finds the leaders of the communities, while also identifying their role in communities.

We present novel indexing and searching schemes for semantic graphs based on the notion of the i.degrees of a node. The i.degrees allow searches performed on the graph to use “type” and connection information, rather than textual labels, to identify nodes. We aim to identify a network graph (fragment) within a large semantic graph (database). A fragment may represent incomplete information that a researcher has collected on a sub-network of interest. While textual labels might be available, they are highly unreliable, and cannot be used for identification of hidden networks. Since this problem comes from the classically NP-hard problem of identifying isomorphic subgraphs, our algorithms are heuristic.

Research on resilient infrastructures in this paper focuses on getting disrupted infrastructures back to normal as soon as possible, i.e., a timely, efficient recovery from disruptions caused by natural, technological and willful disasters. The models and algorithms are presented for the optimal restoration strategy, with consideration of interdependencies of infrastructures. The contribution of this paper will be of assistance to emergency mangers' decision-making processes for mitigating the impact of disasters.

Trust is an important yet complex and little understood aspect of the dyadic relationship between two entities. Trust plays an important role in the formation of coalitions in social networks and in determining how high value of information flows through the network. We present algorithmically quantifiable measures of trust based on communication behavior. We propose that trust results in likely communication behaviors which are statistically different from random communications; detecting these trust-like behaviors allows us to develop a quantitative measure of who trusts whom in the network. We develop algorithms to efficiently compute such behavioral trust and validate these measures on the Twitter network.

In this work, we present the software library graphOnt. The purpose of this library is to automate the process of dynamically extracting ldquointerestingrdquo graphs from semantic networks. Instructions on the extraction are fed into the library via an ontological language specification custom built for this application. A set of SPARQL queries are used to define vertices and edges in the constructed graph. Extracted graphs are returned using the JUNG framework, which offers many algorithmic and visualization options. This work allows a set of individuals analyzing the same semantic network to extract and analyze dynamically created graphs using sophisticated, specific algorithmic tools without needing to manually construct classical graphs from the data.

As sensors become more complex and prevalent, they present their own issues of cost effectiveness and timeliness. It becomes increasingly important to select sensor sets that provide the most information at the least cost and in the most timely and efficient manner. Two typical sensor selection problems appear in a wide range of applications. The first type involves selecting a sensor set that provides the maximum information gain within a budget limit. The other type involves selecting a sensor set that optimizes the tradeoff between information gain and cost. Unfortunately, both require extensive computations due to the exponential search space of sensor subsets. This paper proposes efficient sensor selection algorithms for solving both of these sensor selection problems. The relationships between the sensors and the hypotheses that the sensors aim to assess are modeled with Bayesian networks, and the information gain (benefit) of the sensors with respect to the hypotheses is evaluated by mutual information. We first prove that mutual information is a submodular function in a relaxed condition, which provides theoretical support for the proposed algorithms. For the budget-limit case, we introduce a greedy algorithm that has a constant factor of$(1 - 1/e)$guarantee to the optimal performance. A partitioning procedure is proposed to improve the computational efficiency of the algorithms by efficiently computing mutual information as well as reducing the search space. For the optimal-tradeoff case, a submodular–supermodular procedure is exploited in the proposed algorithm to choose the sensor set that achieves the optimal tradeoff between the benefit and cost in a polynomial-time complexity.

We present a machine learning approach to discover the agent dynamics that drives the evolution of the social groups in a community. We set up the problem by introducing an agent-based hidden Markov model for the agent dynamics: an agent's actions are determined by micro-laws. Nonetheless, We learn the agent dynamics from the observed communications without knowing state transitions. Our approach is to identify the appropriate micro-laws corresponding to an identification of the appropriate parameters in the model. The model identification problem is then formulated as a mixed optimization problem. To solve the problem, we develop a multistage learning process for determining the group structure, the group evolution, and the micro-laws of a community based on the observed set of communications among actors, without knowing the semantic contents. Finally, to test the quality of our approximations and the feasibility of the approach, we present the results of extensive experiments on synthetic data as well as the results on real communities, such as Enron email and Movie newsgroups. Insight into agent dynamics helps us understand the driving forces behind social evolution.

Abstract Trust in information is developing into a vitally important topic as the Internet becomes increasingly ubiquitous within society. Although many discussions of trust in this environment focus on issues like security, technical reliability, or e‐commerce, few address the problem of trust in the information obtained from the Internet. The authors assert that there is a strong need for theoretical and empirical research on trust within the field of information science. As an initial step, the present study develops a model of trust in digital information by integrating the research on trust from the behavioral and social sciences with the research on information quality and human– computer interaction. The model positions trust as a key mediating variable between information quality and information usage, with important consequences for both the producers and consumers of digital information. The authors close by outlining important directions for future research on trust in information science and technology.

We present a modeling laboratory, Virtual Laboratory for the Simulation and Analysis of Social Group Evolution (ViSAGE), that views the organization of human communities and the experience of individuals in a community as contingent upon on the dynamic properties, or micro-laws , of social groups. The laboratory facilitates the theorization and validation of these properties through an iterative research processes that involves (1) forward simulation experiments, which are used to formalize dynamic group properties, (2) reverse engineering from real data on how the parameters are distributed among individual actors in the community, and (3) grounded research, such as participant observation, that follows specific activities of real actors in a community and determines if, or how well, the micro-laws describe the way choices are made in real world, local settings. In this article we report on the design of ViSAGE. We first give some background to the model. Next we detail each component. We then describe a set of simulation experiments that we used to further design and clarify ViSAGE as a tool for studying emergent properties/phenomena in social networks.

The power of social values that helps to surreptitiously shape or formulate our behavior patterns is not only inevitable, but also influential as the directions of our decision making can never seem to escape the impact of this hidden agent. Therefore, the search of such power agent can be validated through a machine learning approach that enables us to discover the agent dynamics in which drives the evolution of the social groups in a community. By doing so, we set up the problem by introducing a parameterized probabilistic model for the agent dynamics: the acts of an agent are determined by micro-laws with unknown parameters. Our approach is to identify the appropriate parameters in the model. To solve the problem, we develop heuristic expectation-maximization style algorithms for determining the micro-laws of a community based on either observed communication links between actors, or the observed evolution of social groups. We present the learning results from the synthetic data as well as the findings on real communities, e.g., Enron email and movie newsgroups.

We present an extended version of a software system SIGHTS (statistical identification of groups hidden in time and space), which can be used for the discovery, analysis, and knowledge visualization of social coalitions in communication networks such as Blog-networks. The evolution of social groups reflects information flow and social dynamics in social networks. Our system discovers such groups by analyzing communication patterns. The goal of SIGHTS is to be an assistant to an analyst in identifying relevant information. The functionality of SIGHTS includes: discovery of coalitions (clusters) and their leaders; finding hidden groups using communication persistence techniques; discovering hidden groups in communication streams; matching topics of the blogs and detecting sentiments; tracking the evolution of clusters; visualization of collections of individual clusters.

Modern society depends on the operations of civil infrastructure systems, such as transportation, energy, telecommunications, and water. These systems have become so interconnected, one relying on another, that disruption of one may lead to disruptions in all. The approach taken in this research is to model these systems by explicitly identifying these interconnections or interdependencies. Definitions of five types of infrastructure inter-dependencies are presented and incorporated into a network flows mathematical representation, i.e., an interdependent layer network model. Using the lower Manhattan region of New York, for illustrative purposes, the implementation of the model is shown. First, the data requirements are presented with realistic data on the interdependent infrastructure systems of power, telecommunications, and subways. Next, a scenario is given that causes major disruption in the services provided by these infrastructures and demonstrates the use of the model in guiding restoration of services. The paper concludes with a discussion of accomplishments and opportunities for future work.

An enduring characteristic of emergencies is the need for near-simultaneous development and deployment of new management procedures. This need can arise with the onset of highly novel problems and the need to act quickly-factors that reduce opportunities for extensive planning in managing the emergency. As a result, decision makers in emergencies must be prepared to improvise. By understanding the cognitive processes in improvisation, organizations can better learn how to plan for, manage, and learn from improvised action. To help create this understanding, this paper reviews and synthesizes prior results on improvisation in the art of jazz, exploring how these results may be applied to improvisation in emergency management. A theory of improvisation in emergency management is then developed and expressed as a cognitive model. The modelpsilas implementation in computer-executable code is then reviewed, along with an illustration of how the model improvises in an emergency situation. Finally, implications of this model and opportunities for future research are presented.

Models are artifacts used to better understand our world.In the context of mining data to create knowledge, the modeler is often faced with discovering and understanding relationships in data that have no apparent analog in the laws of physical science.Much of these data are increasingly abstract; not directly related to the physical, perceivable world.The use of sketches and diagrams as aids in problem solving and as a means of communication is as old as recorded history.Can visualization help us not only to discover the patterns and relationships in these data, but also to use newly discovered knowledge to build computational models?The purpose of this paper is to provide a description of the process of modeling and, based upon theories of cognition, show how visualization can assist in developing and assessing computational models for very large, high dimensional data sets.

Operational risk management is the process of monitoring, evaluating, and changing courses of actions with potential detrimental consequences in real time. In this paper, we extend the decision models proposed in the literature for individual risk managers to account for situations where multiple risk managers are involved. For this purpose, two dynamic and adaptive preference aggregation models for cardinal and ordinal assessments are proposed and discussed. The mechanical aspects of the models are then validated using field data collected from experienced operational risk managers in an individual-expert setting. Sensitivity analysis indicates that the models have enough flexibility to be adapted to account for behavioral considerations. The paper closes with a research agenda.

Eliciting knowledge from multiple experts usually entails the use of groups, and thus is subject to the problems inherent in group dynamics. We present a technique for multiple expert knowledge acquisition that does not rely upon the use of groups and can take advantage of technological advances in communications and computing, i.e., the Internet. The approach uses influence diagrams to represent the individual expert's understanding of the problem situation and develops a Multiple Expert Influence Diagram (MEID), a composite representation of the experts' knowledge. Following a review of present methods for multiple expert knowledge elicitation, we formally define the MEID, describe its manner of construction, and discuss its interpretation. We continue with a review of the issues to be faced in implementation of the technique, and give an illustrative example. Finally, we emphasize the need to provide users of decision aids with defensible measures of the quality of the rules produced by these aids. The MEID-approach is intended to serve as a first step in this direction.

The advent of satellite tracking and communications technology has motivated organizations to investigate centralized real-time support for remote and mobile units and operations. Transportation companies have begun to use satellite tracking systems, production and energy plants are employing intelligent monitoring systems, and on-site emergency managers are being supported by headquarters in their assessment and decision making process. Although different reasoning logics have been proposed for real-time decision making, little attention has been given to the comparative assessment of these approaches. This paper presents empirical evidence on the impact of alternative reasoning logics on individual decision making in real-time. Following the definition of the decision tasks for hazardous operations, commonly used reasoning logics are discussed. Then, results of an empirical comparison of different reasoning approaches are presented. The data were gathered during an experiment with experienced operators. The conclusions of this research are that the selection of the appropriate reasoning logic in support of real-time decision making is crucial.

This paper describes an empirical investigation of the effect of three dimensional (3-D) imaging on group performance in a tactical planning task. The objective of the study is to examine the role that stereoscopic imaging can play in supporting face-to-face group problem solving and decision making-in particular, the alternative generation and evaluation processes in teams. It was hypothesized that with the stereoscopic display, group members would better visualize the information concerning the task environment, producing open communication and information exchanges. The experimental setting was a tactical command and control task, and the quality of the decisions and nature of the group decision process were investigated with three treatments: 1) noncomputerized, i.e., topographic maps with depth cues; 2) two-dimensional (2-D) imaging; and 3) stereoscopic imaging. The results were mixed on group performance. However, those groups with the stereoscopic displays generated more alternatives and spent less time on evaluation. In addition, the stereoscopic decision aid did not interfere with the group problem solving and decision-making processes. The paper concludes with a discussion of potential benefits, and the need to resolve demonstrated weaknesses of the technology.

The need for more effective ways to manage the risk and safety of technological systems has been widely recognized and accepted by government and industry. Traditionally, risk analysis has been considered as part of the process of planning a technological system and addressed the risk inherent in its day-to-day operations. However, risk must also be considered when responding to episodic events whose uniqueness requires taking actions that are variants of, or different from planned operational procedures. The purpose of this paper is to present a new paradigm for real-time risk analysis that capitalizes upon the advances in computer power, human-machine interfaces, and communication technology. A new process for risk assessment and an appropriate reasoning algorithm for choice has been developed which supports the human operator in analyzing risks and making decisions in real-time during unexpected disruptions in the operations of large-scale systems. The process recognizes that although advances in technology may automate many tasks, humans will always be an integral part of managing large-scale systems. The practical realism of the new approach of operational risk management is illustrated by two examples, hazardous material transportation and emergency management. The first example is discussed within the context of a prototype decision support system for interactive real-time risk management.>

Organizational databases are being used to develop rules or guidelines for action that are incorporated into decision processes. Tree induction algorithms of two types, total branching and subset elimination, used in the generation of rules, are reviewed with respect to their treatment of the issue of quality. Based on this assessment, a hybrid approach, probabilistic inductive learning (PrIL), is presented. It provides a probabilistic measure of goodness for an individual rule, enabling the user to set maximum misclassification levels, or minimum reliability levels, with predetermined confidence that each and every rule will satisfy this criterion. The user is able to quantify the reliability of the decision process, i.e., the invoking of the rules, which is of crucial importance in automated decision processes. PrIL and its associated algorithm are described. An illustrative example based on the claims process at a workers' compensation board is presented.>

The authors examine whether one technique for group knowledge elicitation, focus groups, can provide different thoughts and judgments about a problem than individual interviews. The experimental situation was based on the resolution of an ill-structured problem by experienced subjects in a field setting. Videotaped scenarios were used to describe the situation for knowledge acquisition. Ten subjects were used, five for individual interviews and five for the focus groups. It was found that focus groups performed better in generating original responses than individual interviews and that focus groups were at least as good as individual interviews in terms of the quality and acceptability of ideas.>

The philosophical foundations of statistics, operations research, and expert systems are explored, using the notion of metaphor. The authors develop the idea that a root metaphor or a world view exists in both the decision maker and the decision-aid technologies available to solve a problem. To solve a problem correctly, a decision maker must not only use a decision aid correctly, but also first choose the correct decision aid. When an inappropriate decision aid is used, it creates the possibility of so-called type III errors: reaching a wrong solution even though the procedures were followed correctly. As decision aids become both more sophisticated and more user-friendly, they become available for use by more and more naive users, and the possibility of type III errors increases. In addition, as expert systems become more prominent, it is important to be aware of the philosophical biases of the expert and the knowledge engineer as they interact to develop an expert system. A framework is developed that decision makers can use to help them compare the characteristics of various decision aids and choose the most appropriate for solving a particular problem.>

Simulation tests that were carried out to evaluate decision support system design factors in the context of emergency management decision-making are reported. In particular, two basic issues have been examined: the availability of a microcomputer for enhancing data display during an emergency: and the stage or severity of the emergency. The exercise involved the response to a nuclear generating facility accident, a situation that typically would require various inputs from public officials to protect the local community from potential radiation exposure. The results of the experiment indicate: clear support for computerization of emergency management decision support systems; and an apparent need to focus attention on further development of these systems to improve the decision-making capabilities of public managers during the later, less-structured stages of an emergency.

Abstract A reliability theoretic construct is proposed for conceptualizing the process of information flow. It focuses on information produced to satisfy specified purposes or to achieve preconceived objectives. Furthermore, the model incorporates explicitly the concept of an information producer contemplating a choice of action in an uncertain environment. The resulting models are therefore prescriptive in nature. The usefulness of this construct is illustrated by a case analysis of the effectiveness of natural resource data products in land‐use decision making. Measures of system reliability of the information flow network are determined and sensitivity analyses performed. Numerical examples are presented and discussed. The prescriptive nature of this approach permits use of its results to indicate how a data producer can increase the effectiveness of documents by identifying the information flow network, assessing the reliability of each component in the network, finding measures of system reliability, and performing sensitivity analyses to identify the critical components of the system. The result is a closer congruence between the objectives of the data producer and the requirements of users.

A multicriteria optimization approach to the formulation of land use policies for community development is presented. The construct adds to municipal cost-revenue analysis by incorporating socioeconomic and demographic factors. It also deals with community goals in an explicit manner. Results are presented based upon experience in a small city.