Title: Problems faced by students during heavy traffic.
Authors: Prachi Kadam, Rashi Agrawal, Snehal Shinde.
Introduction:
Heavy traffic poses significant challenges to students, affecting various aspects of their academic and personal lives. The unpredictability of heavy traffic makes it challenging for students to maintain a consistent attendance record. Despite their efforts, unavoidable delays caused by traffic congestion often lead to absenteeism, affecting their academic progress and participation in class activities.
Objective:
To understand underlying issues of the problems faced by students during heavy traffic.
1. Object Detection Using Deep Learning Methods in Traffic Scenarios.
The Article discusses object detection for autonomous driving using computer vision and deep learning methods. It talks about traditional computer vision techniques like sliding window and region proposals for object detection that had limitations. Deep learning based single shot detectors like YOLO and SSD were able to directly detect objects in whole images without region proposals, making the detection faster. It classifies different object detection tasks in traffic scenarios like vehicle detection, pedestrian detection, traffic sign detection etc. It mentions some research papers that used deep learning models like Faster R-CNN, YOLO, SSD for detecting objects in traffic images like vehicles, traffic signs. One paper used a Hyper Region Proposal Network (HRPN) to improve small object detection in Faster R-CNN. Another paper obtained three different models based on YOLOv2 for better traffic sign detection as they are small objects. The Article discusses the use of computer vision and deep learning methods for object detection tasks required for autonomous vehicles, and classifies different detection tasks in traffic scenarios. It mentions some related works that used models like Faster R-CNN, YOLO, SSD for detecting vehicles, traffic signs, etc. in images.
2. Application research of urban subway traffic mode based on behavior entropy in the background of big data.
The Article discusses application research of urban subway traffic mode based on behavior entropy in the context of big data. It examines passenger travel behavior in urban subway transportation systems. – It uses historical travel data and behavior entropy theory/methods to study the diversity of passenger travel behavior. This can help describe human behavior well – Models like the urban rail transit operation network station topology model and passenger flow congestion propagation SIR model are used to simulate passenger flow congestion behavior. The Article aims to explore the urban subway transportation mode under the background of big data. It provides a reference basis for subway operation and passenger travel. The results show laws of random fluctuation in passenger flow in the subway train running system. An optimization model of stop time redundancy allocation is also constructed. The Article analyzes subway passenger travel behavior using behavior entropy theories and big data, to better understand transportation patterns and provide insights for subway operations.
Data Collection:
From the above problem we framed five questions to be answered on likert scale (strongly disagree to strongly agree). Data was gathered from the students of KBS. 100 students were surveyed. For each question Mean, Standard Deviation, Standard Error, and tstat was calculated.
Data Analysis:
|
|
Q1. |
Q2 |
Q3. |
Q4. |
Q5. |
|
Mean |
3.91 |
3.52 |
3.24 |
3.68 |
3.76 |
|
Standard Deviation |
1.18 |
1.21 |
1.26 |
1.11 |
1.15 |
|
Standard Error |
0.12 |
0.12 |
0.13 |
0.11 |
0.11 |
|
Tstat |
7.70 |
4.30 |
1.90 |
6.13 |
6.63 |
Q1. At 95% confidence level, tstat is more than 1.96, are positive.
Q2. At 95% confidence level, tstat is more than 1.96, are positive.
Q3. At 95% confidence level, tstat is between 1.96 and -1.96, are neutral.
Q4. At 95% confidence level, tstat is more than 1.96, are positive.
Q5. At 95% confidence level, tstat is more than 1.96, are positive.
Conclusion:
Result No. 1- Students feel anxious about being late to class.
Result No. 2- Students struggle to maintain a consistent attendance record.
Result No. 3- Students are neutral about missing important deadlines.
Result No. 4- Students find it challenging to plan their schedule effectively.
Result No. 5- Students ability to engage in leisure activities or hobbies after class is impacted.
References:
Boukerche, A., & Zhijun Hou. (2022). Object Detection Using Deep Learning Methods in Traffic Scenarios. ACM Computing Surveys, 54(2), 1–35.
Hu, W., Cheng, F., Souri, A., & Chen, M.-Y. (2021). Application research of urban subway traffic mode based on behavior entropy in the background of big data. Journal of High Speed Networks, 27(3), 291–304.