Pouya Mehrannia

Automatic detection of traffic accidents has a crucial effect on improving transportation, public safety, and path planning. Many lives can be saved by the consequent decrease in the time between when the accidents occur and when rescue teams are dispatched, and much travelling time can be saved by notifying drivers to select alternative routes. This problem is challenging mainly because of the rareness of accidents and spatial heterogeneity of the environment. This paper studies deep representation of loop detector data using long‐short term memory (LSTM) network for automatic detection of freeway accidents. The LSTM‐based framework increases class separability in the encoded feature space while reducing the dimension of data. The experiments on real accident and loop detector data collected from the Twin Cities Metro freeways of Minnesota demonstrate that deep representation of traffic flow data using LSTM network has the potential to detect freeway accidents in less than 18 min with a true positive rate of 0.71 and a false positive rate of 0.25 which outperforms other competing methods in the same arrangement.

This paper introduces knowledgebase approximation and fusion using association rules aggregation as a means to facilitate accelerated insight induction from high-dimensional and disparate knowledgebases. There are two typical observations that make approximating knowledgebases of interest: (1) it is quite often that insights can be derived based from a partial set of the samples, and not necessarily from all of them; and (2) generally speaking, it is rare that the knowledge of interest is contained in one knowledgebase, but rather distributed among a disparate set of unidentical knowledgebases. As a matter of fact, the insights derivable from knowledgebases tend to be uncertain, even if they were to be derived from a wholistic analysis of the knowledgebase. Thus, optimal knowledgebase approximation may yield the computational efficiency benefit without necessarily compromising insight accuracy. This paper presents a novel method to approximate a set of knowledgebases based on association rule aggregation using the disjunctive pooling rule. We show that this method can reduce insight discovery time while maintaining approximation accuracy within a desirable level.

Traffic incident detection and localization is an important application in traffic management systems. The ability to detect and localize traffic incidents enables a timely response to accidents and facilitates effective and efficient traffic flow management. This paper presents a sensor-network based approach for tackling the problem of incident localization. Traffic count sensors, which tend to be an element of the road infrastructure, are used as the source of traffic sensory data. Such sensors come in a variety of types and capabilities, providing the potential for complementary and redundant information gathering. Thus, it is conceivable to fuse such sensory information to achieve insightful and accurate incident detection and localization. In this context, the Dempster-Shafer (DS) theory of evidence is used as the foundation for fusing traffic sensory data. In this paper, a traffic model generator and two traffic-counting sensory systems are employed for acquiring traffic data pertinent to the distribution of cars on a given road segment. Experimental analysis on the performance of the proposed approach is provided.

In this research, a multi criteria decision making problem within the Smart Grid has been solved using a modified application of partial order theory (POT), which is an analytical way of dealing with decision making problems. The proposed approach incorporates data of decision matrices as its input. Consequently, by using Monte Carlo simulation, the Ranking Probability Matrix (RPM) is produced which leads us to a total order ranking of the MCDM alternatives. The aim of this paper is to resolve some disadvantages of several conventional methods of determining the best alternative. Thus, the deficiencies of older methods are discussed and their incapability in handling large numbers of alternatives and attributes are resolved. In addition, the uncertainty factor is going to play its role in decision making. We have illustrated the efficiency of our method by implementing it to an IT infrastructure selection problem in Smart Grid.