Passenger flow prediction in bus transportation system using deep learning

Nagaraj, Nandini; Gururaj, H L; Swathi, B H; Hu, Yu‐Chen

doi:10.1007/s11042-022-12306-3

Cited by 41 publications

(14 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The above equation shows that the dot product of x i and x j influences the maximization of the “ a ” value. When the inner product of x i and x is large and from different classes, it forms the margin with maximum width whereas the inner product of the same class does not yield any significance [ 29 ]. The value of w and b can be in turn obtained by obtaining the a -value that maximizes L ( a ).…”

Section: Methodsmentioning

confidence: 99%

Analysis on the Bus Arrival Time Prediction Model for Human-Centric Services Using Data Mining Techniques

Shanthi¹,

Sathishkumar

Babu

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

The human-computer interaction has become inevitable in digital world. HCI helps humans to incorporate technology to resolve even their day-to-day problems. The main objective of the paper is to utilize HCI in Intelligent Transportation Systems. In India, the most common and convenient mode of transportation is the buses. Every state government provides the bus transportation facility to all routes at an affordable cost. The main difficulty faced by the passengers (humans) is lack of information about bus numbers available for the particular route and Estimated Time of Arrival (ETA) of the buses. There may be different reasons for the bus delay. These include heavy traffic, breakdowns, and bad weather conditions. The passengers waiting in the bus stops are neither aware of the delay nor the bus arrival time. These issues can be resolved by providing an HCI-based web/mobile application for the passengers to track their bus locations in real time. They can also check the Estimated Time of Arrival (ETA) of a particular bus, calculated using machine learning techniques by considering the impacts of environmental dynamics, and other factors like traffic density and weather conditions and track their bus locations in real time. This can be achieved by developing a real-time bus management system for the benefit of passengers, bus drivers, and bus managers. This system can effectively address the problems related to bus timing transparency and arrival time forecasting. The buses are equipped with real-time vehicle tracking module containing Raspberry Pi, GPS, and GSM. The traffic density in the current location of the bus and weather data are some of the factors used for the ETA prediction using the Support Vector Regression algorithm. The model showed RMSE of 27 seconds when tested. The model is performing well when compared with other models.

show abstract

Section: Methodsmentioning

confidence: 99%

Analysis on the Bus Arrival Time Prediction Model for Human-Centric Services Using Data Mining Techniques

Shanthi¹,

Sathishkumar

Babu

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…Long short-term memory neural network (LSTM) models are deep learning approaches based on recurrent neural network (RNN) models that are especially well-suited for time series prediction [51]. LSTM models are usually quite accurate [52]; however, their main disadvantage is their complicated training procedure and lack of interpretability. For the lack of interpretability, Monje et al [53] constructed a surrogate tree to obtain rules to explain the LSTM model's black box results.…”

Section: A Forecasting Models For Passenger Load Predictionmentioning

confidence: 99%

Improving the Prediction of Passenger Numbers in Public Transit Networks by Combining Short-Term Forecasts With Real-Time Occupancy Data

Hoppe

Schwinger

Haeger

et al. 2023

IEEE Open J. Intell. Transp. Syst.

View full text Add to dashboard Cite

Passengers of public transportation nowadays expect reliable and accurate travel information. The need for occupancy information is becoming more prevalent in intelligent public transport systems as people started avoiding overcrowded vehicles during the COVID-19 pandemic. Furthermore, public transportation companies require accurate occupancy forecasts to improve service quality. We present a novel approach to improve the prediction of passenger numbers that enhances a day-ahead prediction with real-time data. We first train a baseline predictor on historical automatic passenger counting data. Next, we train a realtime model on the deviations between baseline prediction and observed values, thus capturing events not addressed by the baseline. For the forecast, we attempt to detect emerging patterns in real time and adjust the baseline prediction with deviations from the patterns. Our experiments with data from Germany show that the proposed model improves the forecast of the baseline model and is only outperformed by artificial neural networks in some instances. If the training sets only cover a limited period of up to four months, our approach outperforms competing methods. For larger training sets, there are mixed results in the sense that for some test cases, certain types of neural networks yield slightly better results, but our method still performs well with less training effort, is explainable along the whole prediction process and can be applied to existing prediction methods.

show abstract

“…In recent years, deep learning has shown clear advantages in various networking problems [16][17][18][19][20][21]. People also begin to try deep learning methods in IP geolocation to improve the generalization capabilities.…”

Section: Fine-grained Ip Geolocationmentioning

confidence: 99%

GNN-Geo: A Graph Neural Network-based Fine-grained IP geolocation Framework

Ding¹,

Luo²,

Wang

et al. 2023

IEEE Trans. Netw. Sci. Eng.

View full text Add to dashboard Cite

Rule-based fine-grained IP geolocation methods are hard to generalize in computer networks which do not follow hypothetical rules. Recently, deep learning methods, like multi-layer perceptron (MLP), are tried to increase generalization capabilities. However, MLP is not so suitable for graph-structured data like networks. MLP treats IP addresses as isolated instances and ignores the connection information, which limits geolocation accuracy. In this work, we research how to increase the generalization capability with an emerging graph deep learning method -Graph Neural Network (GNN). First, IP geolocation is re-formulated as an attributed graph node regression problem. Then, we propose a GNN-based IP geolocation framework named GNN-Geo. GNN-Geo consists of a preprocessor, an encoder, messaging passing (MP) layers and a decoder. The preprocessor and encoder transform measurement data into the initial node embeddings. MP layers refine the initial node embeddings by modeling the connection information. The decoder maps the refined embeddings to nodes' locations and relieves the convergence problem by considering prior knowledge. The experiments in 8 real-world IPv4/IPv6 networks in North America, Europe and Asia show the proposed GNN-Geo clearly outperforms the state-of-art rule-based and learning-based baselines. This work verifies the great potential of GNN for fine-grained IP geolocation.

show abstract

Passenger flow prediction in bus transportation system using deep learning

Cited by 41 publications

References 35 publications

Analysis on the Bus Arrival Time Prediction Model for Human-Centric Services Using Data Mining Techniques

Analysis on the Bus Arrival Time Prediction Model for Human-Centric Services Using Data Mining Techniques

Improving the Prediction of Passenger Numbers in Public Transit Networks by Combining Short-Term Forecasts With Real-Time Occupancy Data

GNN-Geo: A Graph Neural Network-based Fine-grained IP geolocation Framework

Contact Info

Product

Resources

About