Design strategies for an artificial neural network based algorithm for automatic incident detection on major arterial streets

Traffic incidents are non-recurring events that can cause a temporary reduction in roadway capacity. They have been recognized as a major contributor to traffic congestion on our nation’s highway systems. To alleviate their impacts on capacity, automatic incident detection (AID) has been applied as an incident management strategy to reduce the total incident duration. AID relies on an algorithm to identify the occurrence of incidents by analyzing real-time traffic data collected from surveillance detectors. Significant research has been performed to develop AID algorithms for incident detection on freeways; however, similar research on major arterial streets remains largely at the initial stage of development and testing. This dissertation research aims to identify design strategies for the deployment of an Artificial Neural Network (ANN) based AID algorithm for major arterial streets. A section of the US-1 corridor in Miami-Dade County, Florida was coded in the CORSIM microscopic simulation model to generate data for both model calibration and validation. To better capture the relationship between the traffic data and the corresponding incident status, Discrete Wavelet Transform (DWT) and data normalization were applied to the simulated data. Multiple ANN models were then developed for different detector configurations, historical data usage, and the selection of traffic flow parameters. To assess the performance of different design alternatives, the model outputs were compared based on both detection rate (DR) and false alarm rate (FAR). The results show that the best models were able to achieve a high DR of between 90% and 95%, a mean time to detect (MTTD) of 55-85 seconds, and a FAR below 4%. The results also show that a detector configuration including only the mid-block and upstream detectors performs almost as well as one that also includes a downstream detector. In addition, DWT was found to be able to improve model performance, and the use of historical data from previous time cycles improved the detection rate. Speed was found to have the most significant impact on the detection rate, while volume was found to contribute the least. The results from this research provide useful insights on the design of AID for arterial street applications.

Design Strategies for an Artificial Neural Network Based Algorithm for Automatic Incident Detection on Major Arterial Streets

Traffic incidents are non-recurring events that can cause a temporary reduction in roadway capacity. They have been recognized as a major contributor to traffic congestion on our national highway systems. To alleviate their impacts on capacity, automatic incident detection (AID) has been applied as an incident management strategy to reduce the total incident duration. AID relies on an algorithm to identify the occurrence of incidents by analyzing real-time traffic data collected from surveillance detectors. Significant research has been performed to develop AID algorithms for incident detection on freeways; however, similar research on major arterial streets remains largely at the initial stage of development and testing. This dissertation research aims to identify design strategies for the deployment of an Artificial Neural Network (ANN) based AID algorithm for major arterial streets. A section of the US-1 corridor in Miami-Dade County, Florida was coded in the CORSIM microscopic simulation model to generate data for both model calibration and validation. To better capture the relationship between the traffic data and the corresponding incident status, Discrete Wavelet Transform (DWT) and data normalization were applied to the simulated data. Multiple ANN models were then developed for different detector configurations, historical data usage, and the selection of traffic flow parameters. To assess the performance of different design alternatives, the model outputs were compared based on both detection rate (DR) and false alarm rate (FAR). The results show that the best models were able to achieve a high DR of between 90% and 95%, a mean time to detect (MTTD) of 55-85 seconds, and a FAR below 4%. The results also show that a detector configuration including only the mid-block and upstream detectors performs almost as well as one that also includes a downstream detector. In addition, DWT was found to be able to improve model performance, and the use of historical data from previous time cycles improved the detection rate. Speed was found to have the most significant impact on the detection rate, while volume was found to contribute the least. The results from this research provide useful insights on the design of AID for arterial street applications.

A case study of the development of the Argentinean community college system

In the 1980's and 1990's, Argentina was undergoing significant political, social and economic changes as a result of the change over from a military driven government to elected governments. A major aspect of the change was an increased emphasis on privatization, and promotion of foreign and domestic investment in Argentina. Higher education leaders were increasingly drawn into developing a national strategy for changing the educational structure to help facilitate changes in other aspects of the society. Preliminary reviews by the Argentinean higher education leaders indicated that adaptation of the American community college promised to help achieve the national goals. ^ The purpose of this study was to determine: if and how the community college concept, an American invention, could be adapted to function in Argentina, a nation with a significantly different history of political, social, cultural and economic development. Achieving this purpose involved: identifying the key leaders in the movement that developed to apply the community college concept in Argentina; the study of their perspectives regarding the movement as it developed; and tracking the assistance given by selected American community college leaders. ^ The case study method was employed in this research, using interview and historical data collection. Key leaders from higher education in the United States and Argentina were interviewed in-depth, to determine their views. An interview protocol with appropriate sub-questions was followed to ensure complete coverage. The interviewees identified several major areas of education in need of change including, the system, access to the system, new areas of study, integration into the hemisphere and, in general, decentralization. Historical review revealed a steady development of the community college concept in Argentina reflected in documentation of events, conceptual writings and legal structures. ^ It was concluded that there is a community college structure beginning to emerge that, so far, in broad outline, follows the structure developed in the United States. It is anticipated however, that future developments will include conceptual aspects to the model reflective of Argentina. ^

Women of African descent: Persistence in completing doctorates

This study examines the educational persistence of women of African descent (WOAD) in pursuit of a doctorate degree at universities in the southeastern United States. WOAD are women of African ancestry born outside the African continent. These women are heirs to an inner dogged determination and spirit to survive despite all odds (Pulliam, 2003, p. 337).This study used Ellis’s (1997) Three Stages for Graduate Student Development as the conceptual framework to examine the persistent strategies used by these women to persist to the completion of their studies. Semi-structured interviews were conducted, transcribed, and analyzed for the 10 study participants. The study examined their beliefs, roles, and support systems utilized during their studies to help propel them to succeed. Interview data were analyzed using cross-case analysis to discover emerging themes and patterns. A comparative analysis was used to analyze the participants’ experiences and discover themes and patterns to review against the historical data. The findings showed that WOAD experienced feelings of isolation, neglect and racial prejudice as doctoral students. Their ability to formulate supportive relationships among each other and in their communities outside the university was key to their persistence to graduation. There still remains a need to create a more engaging and inclusive environment for nontraditional students, particularly those who are WOAD. The study concludes with six strategies of success used by these WOAD to persist to completion. Those six strategies include 1. Keeping the goal of earning a doctorate and graduating foremost on your mind. 2. Set a class and study schedule that allows you to balance family, work and study. 3. Take care of yourself, physically, mentally and emotionally. 4. Keep a goal chart to track your strides toward completion/graduation. 5. Establish a strong support system among your family, friends, colleagues and community; and 6. Do not let the “isms” (racism, sexism, and ageism) deter you.

Phytoplankton bloom status: Chlorophyll a biomass as an indicator of water quality condition in the southern estuaries of Florida, USA

Altered freshwater inflows have affected circulation, salinity, and water quality patterns of Florida Bay, in turn altering the structure and function of this estuary. Changes in water quality and salinity and associated loss of dense turtle grass and other submerged aquatic vegetation (SAV) in Florida Bay have created a condition in the bay where sediments and nutrients have been regularly disturbed, frequently causing large and dense phytoplankton blooms. These algal and cyanobacterial blooms in turn often cause further loss of more recently established SAV, exacerbating the conditions causing the blooms. Chlorophyll a (CHLA) was selected as an indicator of water quality because it is an indicator of phytoplankton biomass, with concentrations reflecting the integrated effect of many of the water quality factors that may be altered by restoration activities. Overall, we assessed the CHLA indicator as being (1) relevant and reflecting the state of the Florida Bay ecosystem, (2) sensitive to ecosystem drivers (stressors, especially nutrient loading), (3) feasible to monitor, and (4) scientifically defensible. Distinct zones within the bay were defined according to statistical and consensual information. Threshold levels of CHLA for each zone were defined using historical data and scientific consensus. A presentation template of condition of the bay using these thresholds is shown as an example of an outreach product.

Essays in economic growth, political economy and institutions

This dissertation analyzes the effects of political and economic institutions on economic development and growth.^ The first essay develops an overlapping-generations political economy model to analyze the incentives of various social groups to finance human capital accumulation through public education expenditures. The contribution of this study to the literature is that it helps explain the observed differences in the economic growth performance of natural resource-abundant countries. The results suggest that the preferred tax rates of the manufacturers on one hand and the political coalition of manufacturers and landowners, on the other hand, are equal to the socially optimal tax rate. However, we show that owners of natural resources prefer an excessively high tax rate, which suppresses aggregate output to a suboptimal level.^ The second essay examines the relationship between the political influence of different social classes and public education spending in panel data estimation. The novel contribution of this paper to the literature is that I proxy the political power and influence of the natural resource owners, manufacturers, and landowners with macroeconomic indicators. The motivation behind this modeling choice is to substantiate the definition of the political power of social classes with economic fundamentals. I use different governance indicators in the estimations to find out how different institutions mediate the overall impact of the political influence of various social classes on public education spending. The results suggest that political stability and absence of violence and rule of law are the important governance indicators.^ The third essay develops a counter argument to Acemoglu et al. (2010) where the thesis is that French institutions and economic reforms fostered economic progress in those German regions invaded by the Napoleonic armies. By providing historical data on urbanization rates used as proxies for economic growth, I demonstrate that similar different rates of economic growth were observed in the regions of France in the post-Napoleonic period as well. The existence of different economic growth rates makes it hard to argue that the differences in economic performance in the German regions that were invaded by the French and those that were spared a similar fate follow from regional differences in economic institutions.^

The contribution of higher education to economic *development in a globalized environment

Amidst concerns about achieving high levels of technology to remain competitive in the global market without compromising economic development, national economies are experiencing a high demand for human capital. As higher education is assumed to be the main source of human capital, this analysis focused on a more specific and less explored area of the generally accepted idea that higher education contributes to economic growth. The purpose of this study, therefore, was to find whether higher education also contributes to economic development, and whether that contribution is more substantial in a globalized context. ^ Consequently, a multiple linear regression analysis was conducted to support with statistical significance the answer to the research question: Does higher education contributes to economic development in the context of globalization? The information analyzed was obtained from historical data of 91 selected countries, and the period of time of the study was 10 years (1990–2000). Some variables, however, were lagged back 5, 10 or 15 years along a 15-year timeframe (1975–1990). The resulting comparative static model was based on the Cobb-Douglas production function and the Solow model to specify economic growth as a function of physical capital, labor, technology, and productivity. Then, formal education, economic development, and globalization were added to the equation. ^ The findings of this study supported the assumption that the independent contribution of the changes in higher education completion and globalization to changes in economic growth is more substantial than the contribution of their interaction. The results also suggested that changes in higher and secondary education completion contribute much more to changes in economic growth in less developed countries than in their more developed counterparts. ^ As a conclusion, based on the results of this study, I proposed the implementation of public policy in less developed countries to promote and expand adequate secondary and higher education systems with the purpose of helping in the achievement of economic development. I also recommended further research efforts on this topic to emphasize the contribution of education to the economy, mainly in less developed countries. ^

Timescales for Detecting a Significant Acceleration in Sea Level Rise

There is observational evidence that global sea level is rising and there is concern that the rate of rise will increase, significantly threatening coastal communities. However, considerable debate remains as to whether the rate of sea level rise is currently increasing and, if so, by how much. Here we provide new insights into sea level accelerations by applying the main methods that have been used previously to search for accelerations in historical data, to identify the timings (with uncertainties) at which accelerations might first be recognized in a statistically significant manner (if not apparent already) in sea level records that we have artificially extended to 2100. We find that the most important approach to earliest possible detection of a significant sea level acceleration lies in improved understanding (and subsequent removal) of interannual to multidecadal variability in sea level records.

Next generation of recommender systems: Algorithms and applications

Personalized recommender systems aim to assist users in retrieving and accessing interesting items by automatically acquiring user preferences from the historical data and matching items with the preferences. In the last decade, recommendation services have gained great attention due to the problem of information overload. However, despite recent advances of personalization techniques, several critical issues in modern recommender systems have not been well studied. These issues include: (1) understanding the accessing patterns of users (i.e., how to effectively model users' accessing behaviors); (2) understanding the relations between users and other objects (i.e., how to comprehensively assess the complex correlations between users and entities in recommender systems); and (3) understanding the interest change of users (i.e., how to adaptively capture users' preference drift over time). To meet the needs of users in modern recommender systems, it is imperative to provide solutions to address the aforementioned issues and apply the solutions to real-world applications. ^ The major goal of this dissertation is to provide integrated recommendation approaches to tackle the challenges of the current generation of recommender systems. In particular, three user-oriented aspects of recommendation techniques were studied, including understanding accessing patterns, understanding complex relations and understanding temporal dynamics. To this end, we made three research contributions. First, we presented various personalized user profiling algorithms to capture click behaviors of users from both coarse- and fine-grained granularities; second, we proposed graph-based recommendation models to describe the complex correlations in a recommender system; third, we studied temporal recommendation approaches in order to capture the preference changes of users, by considering both long-term and short-term user profiles. In addition, a versatile recommendation framework was proposed, in which the proposed recommendation techniques were seamlessly integrated. Different evaluation criteria were implemented in this framework for evaluating recommendation techniques in real-world recommendation applications. ^ In summary, the frequent changes of user interests and item repository lead to a series of user-centric challenges that are not well addressed in the current generation of recommender systems. My work proposed reasonable solutions to these challenges and provided insights on how to address these challenges using a simple yet effective recommendation framework.^

Flood risk analysis of the White Oak Creek watershed in Oak Ridge, Tennessee using MIKE SHE and MIKE 11 software

The purpose of this study was to determine the flooding potential of contaminated areas within the White Oak Creek watershed in the Oak Ridge Reservation in Tennessee. The watershed was analyzed with an integrated surface and subsurface numerical model based on MIKE SHE/MIKE 11 software. The model was calibrated and validated using five decades of historical data. A series of simulations were conducted to determine the watershed response to 25 year, 100 year and 500 year precipitation forecasts; flooding maps were generated for those events. Predicted flood events were compared to Log Pearson III flood flow frequency values for validation. This investigation also provides an improved understanding of the water fluxes between the surface and subsurface subdomains as they affect flood frequencies. In sum, this study presents crucial information to further assess the environmental risks of potential mobilization of contaminants of concern during extreme precipitation events.

Recent Changes in Central and Eastern Pacific El Niño

Recent research indicates that characteristics of El Niño and the Southern Oscillation (ENSO) have changed over the past several decades. Here, I examined different flavors of El Niño in the observational record and the recent changes in the character of El Niño events. The fundamental physical processes that drive ENSO were described and the Eastern Pacific (EP) and Central Pacific (CP) types or flavors of El Niño were defined. Using metrics from the peer-reviewed literature, I examined several historical data sets to interpret El Niño behavior from 1950-2010. A Monte Carlo Simulation was then applied to output from coupled model simulations to test the statistical significance of recent observations surrounding EP and CP El Niño. Results suggested that EP and CP El Niño had been occurring in a similar fashion over the past 60 years with natural variability, but no significant increase in CP El Niño behavior.