LANL Earthquake Prediction

https://www.kaggle.com/c/LANL-Earthquake-Prediction

DataSet

1. Dataset details, such as number of features, instances, data distribution

Features:

Training Data:

• acoustic_data - the seismic signal [int16]
• time_to_failure - the time (in milli seconds) until the next laboratory earthquake [float64]

Training Data instances: 629 million points

Data Distribution:

signal	quaketime
count	1.000000e+07
mean	4.502072e+00
std	1.780707e+01
min	-4.621000e+03
25%	2.000000e+00
50%	4.000000e+00
75%	7.000000e+00
max	3.252000e+03

Test Data:

• seg_id- the test segment ids for which predictions should be made (one prediction per segment)
• acoustic_data - the seismic signal [int16] for which the prediction is made.

Test Data instances: 2624 files, with 150,000 instances for each file => 393,600,000 instances

Techniques we plan to use:

• SVM
• Gradient Boosting
• Random Forests

Experimental methodology:

1. Divide the training data into chunks of 150,000 data points as the test data consists of 150,000 points
2. We are not creating validation dataset as the input dataset is a continguous data from a sensor. Creating validation dataset by choosing the data randomly will not give any good results
3. Scale the data

---

Feature Engineering:

Feature generation: Create several groups of features:

1. Usual aggregations: mean, std, min and max
2. Average difference between the consequitive values in absolute and percent values;
3. Absolute min and max vallues;
4. Aforementioned aggregations for first and last 10000 and 50000 values - I think these data should be useful;
5. Max value to min value and their differencem also count of values bigger that 500 (arbitrary threshold);
6. Quantile features
7. Trend features
8. Rolling features

---

Coding Language:

• Python

---

Team-Members

• Chandra Kiran Saladi ( cxs172130 )
• Shreyash Mane ( ssm170730 )
• Tanya Tukade ( txt171230 )
• Supraja Ponnur ( sxp179130 )