| 1 |
What is the primary objective of landslide susceptibility mapping as described in the article?
|
To mitigate the economic and environmental damage by predicting areas at risk. |
|
Landslide risk mapping is primarily aimed at preventing and mitigating disasters, especially natural disasters such as landslides, which often cause severe damage to life, property, and the environment.
Predicting risk areas allows us to plan for preventive measures in advance, such as building drainage systems, strengthening structures, or evacuating people from dangerous areas, which can significantly reduce potential damage.
Reducing economic and environmental impacts is a consequence of disaster prevention and mitigation, as landslides often cause damage to infrastructure, agriculture, and ecosystems. |
Geology: The study of the physical features of the Earth, such as topography, rocks, soils, and geological processes.
Geographic Information System (GIS): A technology used to collect, analyze, and present geographic data.
Statistics: Analyze quantitative data to find relationships and create models.
Geotechnical Engineering: The application of scientific and engineering principles to solve engineering problems involving soil and rock. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 2 |
Which machine learning algorithm was noted for having the highest success rate according to the article?
|
All of the above equally |
|
Random Forest: เป็นอัลกอริทึมที่ประกอบด้วยต้นไม้การตัดสินใจหลายต้น (Decision Tree) โดยแต่ละต้นจะถูกสร้างขึ้นจากตัวอย่างข้อมูลที่สุ่มเลือกมา ทำให้มีความแม่นยำสูงและสามารถจัดการกับข้อมูลที่มีมิติมากได้ดี
Decision and Regression Tree: ใช้สำหรับทั้งปัญหาการจำแนกประเภทและการทำนายค่าต่อเนื่อง (regression) โดยสร้างกฎการตัดสินใจแบบต้นไม้ |
Overfitting: สถานการณ์ที่โมเดลเรียนรู้ข้อมูลฝึกสอนได้ดีเกินไปจนไม่สามารถนำไปใช้กับข้อมูลใหม่ได้
Underfitting: สถานการณ์ที่โมเดลเรียนรู้ข้อมูลฝึกสอนได้ไม่ดีพอ ทำให้ไม่สามารถจับความสัมพันธ์ในข้อมูลได้
Bias-variance tradeoff: การแลกเปลี่ยนระหว่างความเอนเอียง (bias) และความแปรปรวน (variance) ในโมเดล
Hyperparameter tuning: การปรับค่าพารามิเตอร์ของโมเดลเพื่อให้ได้ประสิทธิภาพที่ดีที่สุด |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 3 |
If the area of Chattogram district is 75% susceptible to landslides, and the highly susceptible zone covers approximately 12% of the district, what is the area (in percentage) that is not highly susceptible?
|
87% |
|
Total area: 100%
High risk area: 12%
Non-high risk areas: 100% - 12% = 87% |
The basic principle of percentages: A percentage is a fraction of a whole number, with a maximum value of 100%.
Principle of Complements: When we know the value of a part of a whole, we can find the remainder by subtracting the known part from the whole. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 4 |
Considering that the total number of analyzed landslides is 255, and 80% were used for training the models, how many landslide instances were used for testing?
|
51 |
|
Total data: We have data on a total of 255 landslides.
Training data: 80% of total data is used to train the model.
Calculate the number of data used for training: 255 x 80% = 204 times.
Test data: The remainder from training is used to test the model.
Therefore, the number of data for testing: 255 - 204 = 51 times. |
Splitting the data into training and test sets is a fundamental step in developing a machine learning system, especially in building a predictive or classification model, such as a landslide prediction model in this case.
Training Set: Used to teach the model to learn patterns and relationships in the data to create a model that can predict outcomes.
Test Set: Used to evaluate the performance of the created model by testing it on data that the model has never seen before and comparing the predicted results of the model with the actual results to measure the accuracy of the model. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 5 |
If the total area of Chattogram district is 7,000 km² and the very high susceptible zone covers 9% of the district, what is the area of the very high susceptible zone in km²?
|
630 km² |
|
The calculation is mathematical: We apply the principle of finding a percentage of a number to find the area.
Complete information provided: The problem specifies the total area and the proportion that is in the very high risk zone, so the answer can be calculated directly. |
Percent: It is a measure of the proportion of a number to a whole number, using the number 100 as the base.
Multiplication: Multiply two numbers together to find the product. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 6 |
Assuming the false positive rate (FPR) for the logistic regression model is 0.05 and the true positive rate (TPR) is 0.95, calculate the specificity of the model.
|
0.95 |
|
Specificity is the ability of the model to correctly identify negative (not positive) samples.
FPR (False Positive Rate) is the proportion of negative samples that are classified as positive by the model.
TPR (True Positive Rate) or Sensitivity is the proportion of Positive samples that are classified as Positive by the model. |
Confusion Matrix: A table showing the results of the model's classification, comparing the predicted values to the actual values.
True Positive (TP), True Negative (TN), False Positive (FP), False Negative (FN): Values in the Confusion Matrix used to calculate model performance metrics such as Accuracy, Precision, Recall, F1-score, Specificity, and Sensitivity.
Binary Classification: A classification that has only two groups (Positive and Negative), which is the type of problem that Logistic Regression is often used for. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 7 |
Given that the area under the ROC curve (AUC) for the logistic regression model is 0.963, and the prediction rate is measured as the area under this curve, rate the model's prediction accuracy.
|
Excellent |
|
What is AUC: AUC stands for Area Under the Curve, which is the area under the ROC (Receiver Operating Characteristic) curve, which is a graph that shows how well a model can distinguish between the positive and negative classes.
What does the AUC mean: The AUC value ranges from 0 to 1.
AUC = 0.5: means the model has no ability to predict any better than random guessing.
AUC = 1: means the model has perfect predictive ability.
Interpretation of AUC values: In general, an AUC value greater than 0.8 is considered a good model, and an AUC value close to 1 indicates a very high predictive accuracy. |
ROC Curve: The ROC curve is a tool used to evaluate the performance of a classification model, considering its True Positive Rate (TPR) and False Positive Rate (FPR).
AUC: The AUC value is a measure of the overall performance of the model, regardless of the threshold selection.
Logistic Regression: It is a classification model used to predict the probability that a given sample will be in the positive class. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 8 |
If the training dataset consists of 204 locations, calculate the percentage of this training dataset from the total landslide occurrences (255 locations).
|
80% |
|
We calculated the proportion directly from the given data by dividing the number of regions of interest (number of training data) by the total number (total data) and converting it to a percentage, which is the standard way to find proportions. |
Proportion: A measure that shows the size of a part relative to the whole, and is usually expressed as a fraction, decimal, or percentage.
Percent: is a ratio expressed as a fraction of 100, used to compare quantities or values. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 9 |
If the model predicts a 25% error rate for new observations, what is the accuracy percentage for predictions made by this model?
|
75% |
|
Accuracy is the proportion of all correct predictions to the total number of predictions.
The error rate is the proportion of all incorrect predictions to the total number of predictions.
So accuracy + error rate = 100%.
If the error rate is 25% then the accuracy is 100% - 25% = 75% |
Probability Theory: The concepts of probability and ratios are used to calculate accuracy and error rates.
Statistics: Measuring model performance is part of statistics, especially in the field of machine learning.
Machine Learning: The concept of model evaluation using metrics such as accuracy, sensitivity, and specificity is fundamental to model development and selection. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 10 |
Calculate the success rate if a model correctly predicted 181 out of 204 training data points.
|
88.73% |
|
From the calculation using the correct formula, the success rate is 88. 73%, which is consistent with the given option.
Calculation method:
Success rate = (number of correctly predicted data / total number of data) x 100%
= (181 / 204) x 100%
≈ 88. 73%
meaning:
This means that the model can correctly predict 88. 73% of the values from all the data used for training. |
Accuracy: A popular measure of a model's performance, it measures the proportion of data that the model correctly predicts compared to the total data.
Model Evaluation: Calculating the accuracy rate is an important step in evaluating the performance of a machine learning model, especially in classification tasks. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 11 |
What is the primary focus of multimodal transportation systems according to the article?
|
Enhancing environmental sustainability and safety. |
|
Diversity of objectives: Multimodal transportation systems do not focus on one factor alone, but rather aim to improve transportation efficiency in many ways, including environmental, safety, and cost.
Global Trends: Today, the world is paying more attention to sustainable development. Reducing environmental impact and increasing transport safety are important goals of modern transport systems.
Limitations of other options:
Minimizing transportation time only: Reducing transportation time is only part of the goal, not the main goal.
Reducing costs irrespective of environmental impact: Reducing costs irrespective of environmental impact can lead to long-term problems.
Focusing solely on increasing transportation capacity: Increasing transportation capacity alone may not be enough if other factors are not taken into account.
Ignoring the risks associated with transportation: Ignoring the risks associated with transportation is impossible. |
Sustainable Logistics: is a concept that focuses on reducing the environmental and social impacts in the transportation process.
Supply Chain: It is a system that connects various activities from production to delivery of goods to consumers. Using multi-modal transportation systems increases the efficiency of the supply chain.
Sustainable Development: A concept that focuses on development that meets the needs of the present without compromising the ability of future generations to meet their own needs. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 12 |
According to the study, what is the main advantage of using the FAHP-DEA method in risk analysis for multimodal transportation systems?
|
It allows for precise risk prioritization and optimization of routes. |
|
FAHP (Analytic Hierarchy Process): is a multi-criteria decision-making method that systematically prioritizes factors based on pairwise comparisons, thus identifying the importance of each factor.
DEA (Data Envelopment Analysis): is a performance analysis method used to evaluate the performance of different departments by comparing them with other departments with similar characteristics, enabling the identification of strengths and weaknesses and finding ways to improve.
Integrating FAHP and DEA: Integrating FAHP and DEA enables a comprehensive integrated risk analysis of public transport systems, prioritizing risks and identifying effective mitigation strategies. |
Decision Theory: FAHP is a methodology based on multi-criteria decision theory, which is a branch of management science.
Efficiency Analysis Theory: DEA is a method based on the theory of efficiency analysis, which is a branch of industrial engineering. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 13 |
If the risk analysis model has five criteria and assigns importance weights such that the total sums up to 1, and the weights for operational risk and security risk are 0.157 and 0.073 respectively, what is the combined weight of the remaining three criteria?
|
0.770 |
|
Weight Principle: In risk analysis, weights are assigned to each criterion to reflect the different levels of importance each criterion has on the overall outcome, with all weights adding up to 1 or 100%.
Calculating the Total Weights: Since we know that the total weight of all five criteria is 1, and we already have weights for the Operational Risk (0. 157) and Security Risk (0. 073), we can find the total weights of the other three criteria by subtracting the total weights from the weights of the two criteria we already know, as follows:
Total weight of the remaining 3 criteria = 1 - 0. 157 - 0. 073 = 0. 770 |
Probability Theory: The concept of importance weights is related to probability theory, where weights can be interpreted as the probability of an event occurring relative to a criterion.
Quantitative Analysis: Assigning importance weights and calculating the total weights is part of quantitative analysis, which is a method that uses numbers and statistics to assess risk. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 14 |
If the probability of an accident occurring on a route is 0.2 and the consequence severity is rated at 0.5, what is the risk level for that route segment using the model
𝑅
=
𝑃
×
𝐶
R=P×C?
|
0.1 |
|
Risk calculation formula: As given in the problem, we use the formula R = P × C, which:
R is the risk level.
P is the probability that the event will occur (Probability)
C is the severity of the impact (Consequence)
Substitute: In this case P = 0. 2 (probability of accident) and C = 0. 5 (severity of impact), so R = 0. 2 × 0. 5 = 0. 1. |
Probability Theory: This is the basis for calculating risk, using a number between 0 and 1 to represent the chance that an event will occur.
Risk Analysis: It is a process used to identify, assess and control potential risks in various activities. The formula R = P × C is one of the basic risk analysis tools. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 15 |
Calculate the aggregate risk score if the weights of the criteria are 0.321, 0.388, 0.157, 0.073, and 0.061, and the local risk scores for a route are 0.5, 0.6, 0.4, 0.3, and 0.2 respectively.
|
0.519 |
|
เนื่องจากการคำนวณตามสูตรที่ได้กล่าวมาข้างต้น ผลลัพธ์ที่ได้คือ 0.519 |
การคำนวณนี้ใช้หลักการถ่วงน้ำหนัก (weighted average) ซึ่งเป็นวิธีการทางสถิติที่นิยมใช้ในการรวมค่าข้อมูลหลายชุดที่มีความสำคัญแตกต่างกัน โดยให้ค่าที่สำคัญกว่ามีน้ำหนักที่มากกว่าในการคำนวณผลลัพธ์สุดท้าย |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 16 |
If the probability assessment for a risk is ranked 3 on a scale of 5 and the severity assessment is also ranked 3, with the transport segment accounting for 20% of the total route distance, calculate the risk assessment using the formula
𝑅
=
𝑃
×
𝐶
×
𝐷
R=P×C×D.
|
1.80 |
|
The calculated risk assessment is 1. 8. The risk assessment formula (R = P × C × D) takes into account both the likelihood (probability) of an event occurring (P) and the potential consequences (severity) of that event (C) ). The distance proportion (D) of the transport segment specifically considers the risk associated with this particular part of the route. |
A higher probability (P) or severity (C) assessment would result in a higher risk assessment (R). The transport segment distance proportion (D) acts as a weighting factor, amplifying the risk if this segment is critical or poses a greater threat. |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 17 |
Given that the weight for environmental risk is 0.061 and the local risk score for a route is 0.4, calculate the contribution of environmental risk to the overall risk score.
|
0.0244 |
|
To calculate the contribution of environmental risks to the overall risk score, we use the principle of weighted average.
Weight: Represents the relative importance of each factor considered, in this case the weight of environmental risk (0. 061).
Score: This is a value that indicates the risk level of that factor, in this case it is the path-specific risk score (0. 4).
Formula:
Support Value = Weight × Score
Calculate:
Support factor = 0. 061 × 0. 4 = 0. 0244
Therefore, the correct answer is 0.0244. |
Weighted Average: This is a method of calculating an average that gives different importance to each data set by multiplying each data set by a given weight and adding all the products together.
Risk Assessment: It is a process used to assess the likelihood and impact of an undesirable event, often used to make decisions about risks in areas such as environmental, health, safety, etc.
|
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 18 |
Calculate the new overall risk score if the weight of infrastructure risk is increased from 0.388 to 0.400 while keeping other parameters constant, given that its local risk score is 0.2.
|
0.160 |
|
ัจจัยอื่นๆ: คำถามระบุว่าปัจจัยอื่นๆ คงที่ แต่ไม่ได้ระบุว่ามีปัจจัยอะไรบ้างและมีน้ำหนักเท่าใด การคำนวณคะแนนความเสี่ยงรวมจำเป็นต้องพิจารณาปัจจัยทั้งหมดที่มีผลต่อความเสี่ยง และน้ำหนักที่กำหนดให้กับแต่ละปัจจัย
สูตรการคำนวณ: ไม่มีการระบุสูตรการคำนวณคะแนนความเสี่ยงรวมที่ใช้ ซึ่งสูตรนี้จะแตกต่างกันไปขึ้นอยู่กับบริบทและวิธีการประเมินความเสี่ยงที่ใช้
ช่วงของคะแนน: ไม่ได้กำหนดช่วงของคะแนนความเสี่ยงที่เป็นไปได้ |
การประเมินความเสี่ยง (Risk Assessment): เป็นกระบวนการที่ใช้ในการระบุ ประเมิน และจัดการความเสี่ยงที่อาจเกิดขึ้น การคำนวณคะแนนความเสี่ยงเป็นส่วนหนึ่งของกระบวนการนี้
น้ำหนักของความเสี่ยง (Weight): เป็นค่าที่แสดงถึงความสำคัญสัมพัทธ์ของแต่ละปัจจัยที่มีต่อความเสี่ยงโดยรวม น้ำหนักจะถูกกำหนดขึ้นอยู่กับความรู้และประสบการณ์ของผู้ประเมิน
คะแนนความเสี่ยงเฉพาะ (Local Risk Score): เป็นคะแนนที่แสดงถึงระดับความเสี่ยงของปัจจัยแต่ละตัว
คะแนนความเสี่ยงรวม (Overall Risk Score): เป็นคะแนนที่แสดงถึงระดับความเสี่ยงโดยรวมของโครงการหรือกิจกรรม |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 19 |
If a mode of transportation has a risk weight of 0.073 and its risk score is reassessed from 0.4 to 0.35, what is the change in its contribution to the overall risk score?
|
0.00365 |
|
การวิเคราะห์ความเสี่ยง (Risk Analysis): เป็นกระบวนการประเมินความน่าจะเป็นและผลกระทบของเหตุการณ์ที่ไม่แน่นอน ซึ่งในที่นี้เราใช้หลักการคูณน้ำหนักความเสี่ยงกับคะแนนความเสี่ยงเพื่อหาส่วนแบ่งความเสี่ยง
การวัดความเสี่ยง (Risk Measurement): การใช้ค่าตัวเลขเพื่อแสดงออกถึงระดับความเสี่ยง ซึ่งในที่นี้เราใช้คะแนนความเสี่ยงเป็นตัวชี้วัด |
การเปลี่ยนแปลงคะแนนความเสี่ยงของวิธีการขนส่งนี้ส่งผลให้ส่วนแบ่งของความเสี่ยงโดยรวมลดลง ซึ่งเป็นผลดีต่อภาพรวมของความเสี่ยงทั้งหมด อย่างไรก็ตาม การลดลงของค่านี้เพียงอย่างเดียวอาจไม่เพียงพอในการตัดสินใจ เนื่องจากยังมีปัจจัยอื่นๆ ที่เกี่ยวข้องกับการประเมินความเสี่ยงโดยรวม |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|
| 20 |
If the local weights of freight-damage risk, infrastructure risk, and operational risk are 0.1, 0.2, and 0.15 respectively, what is their total contribution to the risk score if their respective weights are 0.321, 0.388, and 0.157?
|
0.14647 |
|
น้ำหนักรวมของความเสี่ยง: เพื่อหาค่าน้ำหนักรวมของความเสี่ยงทั้งสาม เราจะนำน้ำหนักที่ปรับปรุงแล้วของแต่ละความเสี่ยงมาบวกกัน |
ทฤษฎีความน่าจะเป็น: ใช้ในการประเมินโอกาสที่เหตุการณ์ความเสี่ยงจะเกิดขึ้น
สถิติ: ใช้ในการวิเคราะห์ข้อมูลเชิงปริมาณและสร้างแบบจำลองเพื่อทำนายความเสี่ยง
การวิเคราะห์ความไว: ใช้ในการประเมินว่าการเปลี่ยนแปลงของปัจจัยหนึ่งๆ จะส่งผลกระทบต่อผลลัพธ์สุดท้ายมากน้อยเพียงใด
การตัดสินใจภายใต้ความไม่แน่นอน: ใช้ในการตัดสินใจเลือกทางเลือกที่เหมาะสมที่สุดในสถานการณ์ที่มีความไม่แน่นอน |
7 |
-.50
-.25
+.25
เต็ม
0
-35%
+30%
+35%
|