A. Rahimi-kian - Academia.edu (original) (raw)

Papers by A. Rahimi-kian

2012 IEEE International Conference on Industrial Technology, 2012

Today, Traffic flow forecast, as one of the topics in intelligent transportation systems, plays a... more Today, Traffic flow forecast, as one of the topics in intelligent transportation systems, plays an important role in areas such as controlling traffic flow, traffic lights and travel time. This paper, in accordance with the power of intelligent systems in modeling and forecasting, uses MLP neural network in traffic forecast and compares MLP results with the conventional methods. As the behavioral patterns of traffic flow are different, utilizing this model as a nonlinear one and adoptable with the environmental conditions is very convenient. Since 'Mutual Information' is a great tool to calculate the interdependence of previous data, nonlinear interdependence of the data can be easily obtained using this tool. This paper has tried to show the increase in the precision of traffic flow forecast alongside the decrease in the amount of calculations even under the circumstances of data loss or data damage in various weather conditions like precipitations and also different traffic problems like congestion or accidents. This is achieved utilizing the algorithm of calculating the mutuality of the previous data and choosing the best for the prediction purposes while is evaluated using the real traffic data with different chaotic modes.

2012 IEEE International Conference on Industrial Technology, 2012

Today, Traffic flow forecast, as one of the topics in intelligent transportation systems, plays a... more Today, Traffic flow forecast, as one of the topics in intelligent transportation systems, plays an important role in areas such as controlling traffic flow, traffic lights and travel time. This paper, in accordance with the power of intelligent systems in modeling and forecasting, uses MLP neural network in traffic forecast and compares MLP results with the conventional methods. As the behavioral patterns of traffic flow are different, utilizing this model as a nonlinear one and adoptable with the environmental conditions is very convenient. Since 'Mutual Information' is a great tool to calculate the interdependence of previous data, nonlinear interdependence of the data can be easily obtained using this tool. This paper has tried to show the increase in the precision of traffic flow forecast alongside the decrease in the amount of calculations even under the circumstances of data loss or data damage in various weather conditions like precipitations and also different traffic problems like congestion or accidents. This is achieved utilizing the algorithm of calculating the mutuality of the previous data and choosing the best for the prediction purposes while is evaluated using the real traffic data with different chaotic modes. and is a real-time, completely nonlinear, high dimensional and non-stationary stochastic process . Traffic flow measurement and forecasting, appropriately addressed in the past two decades, has numerous benefits for the design and operation of highways. Traffic flow forecasting in a highway network is defined as the algorithm-based prediction of all traffic network variables on the basis of existing traffic . Most studies on forecasting traffic flow have been conducted on highways (and freeways) and only some on street settings. Traffic flow forecasting is mainly classified into two categories: long-term and short-term predictions. In short-term prediction, traffic flow is predicted in the next moments (typically between 5 to 30 minutes) on the basis of real-time online or historical data. Of course for a shorter term prediction, online data are adequate. However, for terms longer than 20 minutes historical data are essential. In long-term prediction, traffic flow is predicted for terms longer than 30 minutes, hours, days, weeks, months and even years. Traffic conditions may vary from one day to another or even from one moment to another. This variation may be due to changeable weather conditions, road accidents, cultural or political occasions and events, the types of vehicles and driver characteristics. These variables affect both short-term and long-term prediction efforts and processes and may also affect prediction accuracy for traffic flow [5].

2012 IEEE International Conference on Industrial Technology, 2012

Today, Traffic flow forecast, as one of the topics in intelligent transportation systems, plays a... more Today, Traffic flow forecast, as one of the topics in intelligent transportation systems, plays an important role in areas such as controlling traffic flow, traffic lights and travel time. This paper, in accordance with the power of intelligent systems in modeling and forecasting, uses MLP neural network in traffic forecast and compares MLP results with the conventional methods. As the behavioral patterns of traffic flow are different, utilizing this model as a nonlinear one and adoptable with the environmental conditions is very convenient. Since 'Mutual Information' is a great tool to calculate the interdependence of previous data, nonlinear interdependence of the data can be easily obtained using this tool. This paper has tried to show the increase in the precision of traffic flow forecast alongside the decrease in the amount of calculations even under the circumstances of data loss or data damage in various weather conditions like precipitations and also different traffic problems like congestion or accidents. This is achieved utilizing the algorithm of calculating the mutuality of the previous data and choosing the best for the prediction purposes while is evaluated using the real traffic data with different chaotic modes.

2012 IEEE International Conference on Industrial Technology, 2012

Today, Traffic flow forecast, as one of the topics in intelligent transportation systems, plays a... more Today, Traffic flow forecast, as one of the topics in intelligent transportation systems, plays an important role in areas such as controlling traffic flow, traffic lights and travel time. This paper, in accordance with the power of intelligent systems in modeling and forecasting, uses MLP neural network in traffic forecast and compares MLP results with the conventional methods. As the behavioral patterns of traffic flow are different, utilizing this model as a nonlinear one and adoptable with the environmental conditions is very convenient. Since 'Mutual Information' is a great tool to calculate the interdependence of previous data, nonlinear interdependence of the data can be easily obtained using this tool. This paper has tried to show the increase in the precision of traffic flow forecast alongside the decrease in the amount of calculations even under the circumstances of data loss or data damage in various weather conditions like precipitations and also different traffic problems like congestion or accidents. This is achieved utilizing the algorithm of calculating the mutuality of the previous data and choosing the best for the prediction purposes while is evaluated using the real traffic data with different chaotic modes. and is a real-time, completely nonlinear, high dimensional and non-stationary stochastic process . Traffic flow measurement and forecasting, appropriately addressed in the past two decades, has numerous benefits for the design and operation of highways. Traffic flow forecasting in a highway network is defined as the algorithm-based prediction of all traffic network variables on the basis of existing traffic . Most studies on forecasting traffic flow have been conducted on highways (and freeways) and only some on street settings. Traffic flow forecasting is mainly classified into two categories: long-term and short-term predictions. In short-term prediction, traffic flow is predicted in the next moments (typically between 5 to 30 minutes) on the basis of real-time online or historical data. Of course for a shorter term prediction, online data are adequate. However, for terms longer than 20 minutes historical data are essential. In long-term prediction, traffic flow is predicted for terms longer than 30 minutes, hours, days, weeks, months and even years. Traffic conditions may vary from one day to another or even from one moment to another. This variation may be due to changeable weather conditions, road accidents, cultural or political occasions and events, the types of vehicles and driver characteristics. These variables affect both short-term and long-term prediction efforts and processes and may also affect prediction accuracy for traffic flow [5].

IAS '97. Conference Record of the 1997 IEEE Industry Applications Conference Thirty-Second IAS Annual Meeting, 2000

IAS '97. Conference Record of the 1997 IEEE Industry Applications Conference Thirty-Second IAS Annual Meeting, 2000

IAS '97. Conference Record of the 1997 IEEE Industry Applications Conference Thirty-Second IAS Annual Meeting, 2000

IAS '97. Conference Record of the 1997 IEEE Industry Applications Conference Thirty-Second IAS Annual Meeting, 2000

2012 IEEE PES Innovative Smart Grid Technologies (ISGT), 2012

The user has requested enhancement of the downloaded file. All in-text references underlined in b... more The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document and are linked to publications on ResearchGate, letting you access and read them immediately.

2012 IEEE PES Innovative Smart Grid Technologies (ISGT), 2012

The user has requested enhancement of the downloaded file. All in-text references underlined in b... more The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document and are linked to publications on ResearchGate, letting you access and read them immediately.

2012 IEEE PES Innovative Smart Grid Technologies (ISGT), 2012

The user has requested enhancement of the downloaded file. All in-text references underlined in b... more The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document and are linked to publications on ResearchGate, letting you access and read them immediately.

2012 IEEE PES Innovative Smart Grid Technologies (ISGT), 2012

The user has requested enhancement of the downloaded file. All in-text references underlined in b... more The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document and are linked to publications on ResearchGate, letting you access and read them immediately.

2012 25th IEEE Canadian Conference on Electrical and Computer Engineering (CCECE), 2012

Traffic speed prediction is an important problem in the research area of intelligent transportati... more Traffic speed prediction is an important problem in the research area of intelligent transportation system (ITS). Recently, artificial neural networks models, such as MLP, have been used in various applications over nonlinear time series prediction such as traffic control. In modeling, irrelevant inputs cause the deterioration of performance and an increase in calculation cost. Therefore, to have an accurate model, some strategies are needed to choose a set of most relevant inputs. Mutual information (MI) is very effective in evaluating the nonlinear relevance of each input from the view of information theory. Feature selection (FS) method is an improved version of the MI technique. This paper presents a novel traffic speed prediction model using MLP predictor and MIFS algorithm. Performance of the proposed MIFS algorithm and MLP predictor is evaluated via simulations using MATLAB subroutine. To validate the algorithm, traffic data of Minnesota highways is used.

2012 25th IEEE Canadian Conference on Electrical and Computer Engineering (CCECE), 2012

Traffic speed prediction is an important problem in the research area of intelligent transportati... more Traffic speed prediction is an important problem in the research area of intelligent transportation system (ITS). Recently, artificial neural networks models, such as MLP, have been used in various applications over nonlinear time series prediction such as traffic control. In modeling, irrelevant inputs cause the deterioration of performance and an increase in calculation cost. Therefore, to have an accurate model, some strategies are needed to choose a set of most relevant inputs. Mutual information (MI) is very effective in evaluating the nonlinear relevance of each input from the view of information theory. Feature selection (FS) method is an improved version of the MI technique. This paper presents a novel traffic speed prediction model using MLP predictor and MIFS algorithm. Performance of the proposed MIFS algorithm and MLP predictor is evaluated via simulations using MATLAB subroutine. To validate the algorithm, traffic data of Minnesota highways is used.

2012 25th IEEE Canadian Conference on Electrical and Computer Engineering (CCECE), 2012

Traffic speed prediction is an important problem in the research area of intelligent transportati... more Traffic speed prediction is an important problem in the research area of intelligent transportation system (ITS). Recently, artificial neural networks models, such as MLP, have been used in various applications over nonlinear time series prediction such as traffic control. In modeling, irrelevant inputs cause the deterioration of performance and an increase in calculation cost. Therefore, to have an accurate model, some strategies are needed to choose a set of most relevant inputs. Mutual information (MI) is very effective in evaluating the nonlinear relevance of each input from the view of information theory. Feature selection (FS) method is an improved version of the MI technique. This paper presents a novel traffic speed prediction model using MLP predictor and MIFS algorithm. Performance of the proposed MIFS algorithm and MLP predictor is evaluated via simulations using MATLAB subroutine. To validate the algorithm, traffic data of Minnesota highways is used.

2012 25th IEEE Canadian Conference on Electrical and Computer Engineering (CCECE), 2012

Traffic speed prediction is an important problem in the research area of intelligent transportati... more Traffic speed prediction is an important problem in the research area of intelligent transportation system (ITS). Recently, artificial neural networks models, such as MLP, have been used in various applications over nonlinear time series prediction such as traffic control. In modeling, irrelevant inputs cause the deterioration of performance and an increase in calculation cost. Therefore, to have an accurate model, some strategies are needed to choose a set of most relevant inputs. Mutual information (MI) is very effective in evaluating the nonlinear relevance of each input from the view of information theory. Feature selection (FS) method is an improved version of the MI technique. This paper presents a novel traffic speed prediction model using MLP predictor and MIFS algorithm. Performance of the proposed MIFS algorithm and MLP predictor is evaluated via simulations using MATLAB subroutine. To validate the algorithm, traffic data of Minnesota highways is used.

EUROCON 2005 - The International Conference on "Computer as a Tool", 2005

Particle Swarm Optimization, a nature-inspired evolutionary algorithm, has been successful in sol... more Particle Swarm Optimization, a nature-inspired evolutionary algorithm, has been successful in solving a wide range of real-value optimization problems. However, little attempts have been made to extend it to discrete problems. In this paper, a new binary particle swarm optimization method based on the theory of immunity in biology is proposed. In spite of the simplicity of the technique, simulation results show the improvement of the searching ability and increment in the convergence speed in comparison with the other binary PSO and genetic algorithm.

EUROCON 2005 - The International Conference on "Computer as a Tool", 2005

Particle Swarm Optimization, a nature-inspired evolutionary algorithm, has been successful in sol... more Particle Swarm Optimization, a nature-inspired evolutionary algorithm, has been successful in solving a wide range of real-value optimization problems. However, little attempts have been made to extend it to discrete problems. In this paper, a new binary particle swarm optimization method based on the theory of immunity in biology is proposed. In spite of the simplicity of the technique, simulation results show the improvement of the searching ability and increment in the convergence speed in comparison with the other binary PSO and genetic algorithm.

EUROCON 2005 - The International Conference on "Computer as a Tool", 2005

Particle Swarm Optimization, a nature-inspired evolutionary algorithm, has been successful in sol... more Particle Swarm Optimization, a nature-inspired evolutionary algorithm, has been successful in solving a wide range of real-value optimization problems. However, little attempts have been made to extend it to discrete problems. In this paper, a new binary particle swarm optimization method based on the theory of immunity in biology is proposed. In spite of the simplicity of the technique, simulation results show the improvement of the searching ability and increment in the convergence speed in comparison with the other binary PSO and genetic algorithm.

EUROCON 2005 - The International Conference on "Computer as a Tool", 2005

Particle Swarm Optimization, a nature-inspired evolutionary algorithm, has been successful in sol... more Particle Swarm Optimization, a nature-inspired evolutionary algorithm, has been successful in solving a wide range of real-value optimization problems. However, little attempts have been made to extend it to discrete problems. In this paper, a new binary particle swarm optimization method based on the theory of immunity in biology is proposed. In spite of the simplicity of the technique, simulation results show the improvement of the searching ability and increment in the convergence speed in comparison with the other binary PSO and genetic algorithm.

2012 IEEE International Conference on Industrial Technology, 2012

Today, Traffic flow forecast, as one of the topics in intelligent transportation systems, plays a... more Today, Traffic flow forecast, as one of the topics in intelligent transportation systems, plays an important role in areas such as controlling traffic flow, traffic lights and travel time. This paper, in accordance with the power of intelligent systems in modeling and forecasting, uses MLP neural network in traffic forecast and compares MLP results with the conventional methods. As the behavioral patterns of traffic flow are different, utilizing this model as a nonlinear one and adoptable with the environmental conditions is very convenient. Since 'Mutual Information' is a great tool to calculate the interdependence of previous data, nonlinear interdependence of the data can be easily obtained using this tool. This paper has tried to show the increase in the precision of traffic flow forecast alongside the decrease in the amount of calculations even under the circumstances of data loss or data damage in various weather conditions like precipitations and also different traffic problems like congestion or accidents. This is achieved utilizing the algorithm of calculating the mutuality of the previous data and choosing the best for the prediction purposes while is evaluated using the real traffic data with different chaotic modes.

2012 IEEE International Conference on Industrial Technology, 2012

Today, Traffic flow forecast, as one of the topics in intelligent transportation systems, plays a... more Today, Traffic flow forecast, as one of the topics in intelligent transportation systems, plays an important role in areas such as controlling traffic flow, traffic lights and travel time. This paper, in accordance with the power of intelligent systems in modeling and forecasting, uses MLP neural network in traffic forecast and compares MLP results with the conventional methods. As the behavioral patterns of traffic flow are different, utilizing this model as a nonlinear one and adoptable with the environmental conditions is very convenient. Since 'Mutual Information' is a great tool to calculate the interdependence of previous data, nonlinear interdependence of the data can be easily obtained using this tool. This paper has tried to show the increase in the precision of traffic flow forecast alongside the decrease in the amount of calculations even under the circumstances of data loss or data damage in various weather conditions like precipitations and also different traffic problems like congestion or accidents. This is achieved utilizing the algorithm of calculating the mutuality of the previous data and choosing the best for the prediction purposes while is evaluated using the real traffic data with different chaotic modes. and is a real-time, completely nonlinear, high dimensional and non-stationary stochastic process . Traffic flow measurement and forecasting, appropriately addressed in the past two decades, has numerous benefits for the design and operation of highways. Traffic flow forecasting in a highway network is defined as the algorithm-based prediction of all traffic network variables on the basis of existing traffic . Most studies on forecasting traffic flow have been conducted on highways (and freeways) and only some on street settings. Traffic flow forecasting is mainly classified into two categories: long-term and short-term predictions. In short-term prediction, traffic flow is predicted in the next moments (typically between 5 to 30 minutes) on the basis of real-time online or historical data. Of course for a shorter term prediction, online data are adequate. However, for terms longer than 20 minutes historical data are essential. In long-term prediction, traffic flow is predicted for terms longer than 30 minutes, hours, days, weeks, months and even years. Traffic conditions may vary from one day to another or even from one moment to another. This variation may be due to changeable weather conditions, road accidents, cultural or political occasions and events, the types of vehicles and driver characteristics. These variables affect both short-term and long-term prediction efforts and processes and may also affect prediction accuracy for traffic flow [5].

2012 IEEE International Conference on Industrial Technology, 2012

Today, Traffic flow forecast, as one of the topics in intelligent transportation systems, plays a... more Today, Traffic flow forecast, as one of the topics in intelligent transportation systems, plays an important role in areas such as controlling traffic flow, traffic lights and travel time. This paper, in accordance with the power of intelligent systems in modeling and forecasting, uses MLP neural network in traffic forecast and compares MLP results with the conventional methods. As the behavioral patterns of traffic flow are different, utilizing this model as a nonlinear one and adoptable with the environmental conditions is very convenient. Since 'Mutual Information' is a great tool to calculate the interdependence of previous data, nonlinear interdependence of the data can be easily obtained using this tool. This paper has tried to show the increase in the precision of traffic flow forecast alongside the decrease in the amount of calculations even under the circumstances of data loss or data damage in various weather conditions like precipitations and also different traffic problems like congestion or accidents. This is achieved utilizing the algorithm of calculating the mutuality of the previous data and choosing the best for the prediction purposes while is evaluated using the real traffic data with different chaotic modes.

2012 IEEE International Conference on Industrial Technology, 2012

Today, Traffic flow forecast, as one of the topics in intelligent transportation systems, plays a... more Today, Traffic flow forecast, as one of the topics in intelligent transportation systems, plays an important role in areas such as controlling traffic flow, traffic lights and travel time. This paper, in accordance with the power of intelligent systems in modeling and forecasting, uses MLP neural network in traffic forecast and compares MLP results with the conventional methods. As the behavioral patterns of traffic flow are different, utilizing this model as a nonlinear one and adoptable with the environmental conditions is very convenient. Since 'Mutual Information' is a great tool to calculate the interdependence of previous data, nonlinear interdependence of the data can be easily obtained using this tool. This paper has tried to show the increase in the precision of traffic flow forecast alongside the decrease in the amount of calculations even under the circumstances of data loss or data damage in various weather conditions like precipitations and also different traffic problems like congestion or accidents. This is achieved utilizing the algorithm of calculating the mutuality of the previous data and choosing the best for the prediction purposes while is evaluated using the real traffic data with different chaotic modes. and is a real-time, completely nonlinear, high dimensional and non-stationary stochastic process . Traffic flow measurement and forecasting, appropriately addressed in the past two decades, has numerous benefits for the design and operation of highways. Traffic flow forecasting in a highway network is defined as the algorithm-based prediction of all traffic network variables on the basis of existing traffic . Most studies on forecasting traffic flow have been conducted on highways (and freeways) and only some on street settings. Traffic flow forecasting is mainly classified into two categories: long-term and short-term predictions. In short-term prediction, traffic flow is predicted in the next moments (typically between 5 to 30 minutes) on the basis of real-time online or historical data. Of course for a shorter term prediction, online data are adequate. However, for terms longer than 20 minutes historical data are essential. In long-term prediction, traffic flow is predicted for terms longer than 30 minutes, hours, days, weeks, months and even years. Traffic conditions may vary from one day to another or even from one moment to another. This variation may be due to changeable weather conditions, road accidents, cultural or political occasions and events, the types of vehicles and driver characteristics. These variables affect both short-term and long-term prediction efforts and processes and may also affect prediction accuracy for traffic flow [5].

IAS '97. Conference Record of the 1997 IEEE Industry Applications Conference Thirty-Second IAS Annual Meeting, 2000

IAS '97. Conference Record of the 1997 IEEE Industry Applications Conference Thirty-Second IAS Annual Meeting, 2000

IAS '97. Conference Record of the 1997 IEEE Industry Applications Conference Thirty-Second IAS Annual Meeting, 2000

IAS '97. Conference Record of the 1997 IEEE Industry Applications Conference Thirty-Second IAS Annual Meeting, 2000

2012 IEEE PES Innovative Smart Grid Technologies (ISGT), 2012

The user has requested enhancement of the downloaded file. All in-text references underlined in b... more The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document and are linked to publications on ResearchGate, letting you access and read them immediately.

2012 IEEE PES Innovative Smart Grid Technologies (ISGT), 2012

The user has requested enhancement of the downloaded file. All in-text references underlined in b... more The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document and are linked to publications on ResearchGate, letting you access and read them immediately.

2012 IEEE PES Innovative Smart Grid Technologies (ISGT), 2012

The user has requested enhancement of the downloaded file. All in-text references underlined in b... more The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document and are linked to publications on ResearchGate, letting you access and read them immediately.

2012 IEEE PES Innovative Smart Grid Technologies (ISGT), 2012

The user has requested enhancement of the downloaded file. All in-text references underlined in b... more The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document and are linked to publications on ResearchGate, letting you access and read them immediately.

2012 25th IEEE Canadian Conference on Electrical and Computer Engineering (CCECE), 2012

Traffic speed prediction is an important problem in the research area of intelligent transportati... more Traffic speed prediction is an important problem in the research area of intelligent transportation system (ITS). Recently, artificial neural networks models, such as MLP, have been used in various applications over nonlinear time series prediction such as traffic control. In modeling, irrelevant inputs cause the deterioration of performance and an increase in calculation cost. Therefore, to have an accurate model, some strategies are needed to choose a set of most relevant inputs. Mutual information (MI) is very effective in evaluating the nonlinear relevance of each input from the view of information theory. Feature selection (FS) method is an improved version of the MI technique. This paper presents a novel traffic speed prediction model using MLP predictor and MIFS algorithm. Performance of the proposed MIFS algorithm and MLP predictor is evaluated via simulations using MATLAB subroutine. To validate the algorithm, traffic data of Minnesota highways is used.

2012 25th IEEE Canadian Conference on Electrical and Computer Engineering (CCECE), 2012

Traffic speed prediction is an important problem in the research area of intelligent transportati... more Traffic speed prediction is an important problem in the research area of intelligent transportation system (ITS). Recently, artificial neural networks models, such as MLP, have been used in various applications over nonlinear time series prediction such as traffic control. In modeling, irrelevant inputs cause the deterioration of performance and an increase in calculation cost. Therefore, to have an accurate model, some strategies are needed to choose a set of most relevant inputs. Mutual information (MI) is very effective in evaluating the nonlinear relevance of each input from the view of information theory. Feature selection (FS) method is an improved version of the MI technique. This paper presents a novel traffic speed prediction model using MLP predictor and MIFS algorithm. Performance of the proposed MIFS algorithm and MLP predictor is evaluated via simulations using MATLAB subroutine. To validate the algorithm, traffic data of Minnesota highways is used.

2012 25th IEEE Canadian Conference on Electrical and Computer Engineering (CCECE), 2012

Traffic speed prediction is an important problem in the research area of intelligent transportati... more Traffic speed prediction is an important problem in the research area of intelligent transportation system (ITS). Recently, artificial neural networks models, such as MLP, have been used in various applications over nonlinear time series prediction such as traffic control. In modeling, irrelevant inputs cause the deterioration of performance and an increase in calculation cost. Therefore, to have an accurate model, some strategies are needed to choose a set of most relevant inputs. Mutual information (MI) is very effective in evaluating the nonlinear relevance of each input from the view of information theory. Feature selection (FS) method is an improved version of the MI technique. This paper presents a novel traffic speed prediction model using MLP predictor and MIFS algorithm. Performance of the proposed MIFS algorithm and MLP predictor is evaluated via simulations using MATLAB subroutine. To validate the algorithm, traffic data of Minnesota highways is used.

2012 25th IEEE Canadian Conference on Electrical and Computer Engineering (CCECE), 2012

Traffic speed prediction is an important problem in the research area of intelligent transportati... more Traffic speed prediction is an important problem in the research area of intelligent transportation system (ITS). Recently, artificial neural networks models, such as MLP, have been used in various applications over nonlinear time series prediction such as traffic control. In modeling, irrelevant inputs cause the deterioration of performance and an increase in calculation cost. Therefore, to have an accurate model, some strategies are needed to choose a set of most relevant inputs. Mutual information (MI) is very effective in evaluating the nonlinear relevance of each input from the view of information theory. Feature selection (FS) method is an improved version of the MI technique. This paper presents a novel traffic speed prediction model using MLP predictor and MIFS algorithm. Performance of the proposed MIFS algorithm and MLP predictor is evaluated via simulations using MATLAB subroutine. To validate the algorithm, traffic data of Minnesota highways is used.

EUROCON 2005 - The International Conference on "Computer as a Tool", 2005

Particle Swarm Optimization, a nature-inspired evolutionary algorithm, has been successful in sol... more Particle Swarm Optimization, a nature-inspired evolutionary algorithm, has been successful in solving a wide range of real-value optimization problems. However, little attempts have been made to extend it to discrete problems. In this paper, a new binary particle swarm optimization method based on the theory of immunity in biology is proposed. In spite of the simplicity of the technique, simulation results show the improvement of the searching ability and increment in the convergence speed in comparison with the other binary PSO and genetic algorithm.

EUROCON 2005 - The International Conference on "Computer as a Tool", 2005

Particle Swarm Optimization, a nature-inspired evolutionary algorithm, has been successful in sol... more Particle Swarm Optimization, a nature-inspired evolutionary algorithm, has been successful in solving a wide range of real-value optimization problems. However, little attempts have been made to extend it to discrete problems. In this paper, a new binary particle swarm optimization method based on the theory of immunity in biology is proposed. In spite of the simplicity of the technique, simulation results show the improvement of the searching ability and increment in the convergence speed in comparison with the other binary PSO and genetic algorithm.

EUROCON 2005 - The International Conference on "Computer as a Tool", 2005

Particle Swarm Optimization, a nature-inspired evolutionary algorithm, has been successful in sol... more Particle Swarm Optimization, a nature-inspired evolutionary algorithm, has been successful in solving a wide range of real-value optimization problems. However, little attempts have been made to extend it to discrete problems. In this paper, a new binary particle swarm optimization method based on the theory of immunity in biology is proposed. In spite of the simplicity of the technique, simulation results show the improvement of the searching ability and increment in the convergence speed in comparison with the other binary PSO and genetic algorithm.

EUROCON 2005 - The International Conference on "Computer as a Tool", 2005

Particle Swarm Optimization, a nature-inspired evolutionary algorithm, has been successful in sol... more Particle Swarm Optimization, a nature-inspired evolutionary algorithm, has been successful in solving a wide range of real-value optimization problems. However, little attempts have been made to extend it to discrete problems. In this paper, a new binary particle swarm optimization method based on the theory of immunity in biology is proposed. In spite of the simplicity of the technique, simulation results show the improvement of the searching ability and increment in the convergence speed in comparison with the other binary PSO and genetic algorithm.