This document discusses using a random forest algorithm to predict flight prices. It describes the requirements, including hardware of a quad core Intel Core i3 and 4 GB of RAM, and software including Anaconda Navigator and Python. The random forest algorithm is used due to its ability to handle both classification and regression problems. By collecting and preprocessing a dataset and applying the random forest model, the results showed better accuracy than other algorithms for predicting flight prices based on attributes like duration, arrival time, price, source and destination.

1. FLIGHT PRICE PREDICTION

2. Contents
• Introduction
• Requirements
• Algorithm Used
• Conclusion
• References

3. Introduction
1. Perfect time for purchasing plane ticket by the passenger’s view is difficult passengers get very less
information of future business price rates.
2. Airlines use different strategies of pricing for their tickets, later taking the decision on price because order
shows higher value for the approximation models.
3. The causes behind the difficult system is each Planes has limited number of seats to be filed, so airlines must
regulate demand.
4. Duration, Arrival time, Price, Source, Destination and much more these are the attribute used for flight price
prediction.

4. Requirements
• Hardware Requirements
• Quad core Intel Core i3
• 4GB of RAM

5. Software Requirements
• Anaconda navigator
• Python

6. Algorithm Used
• Random Forest Algorithm
• One of the popular machine learning algorithms which belongs to the supervised learning technique is
Random Forest.
• Process of combining multiple classifiers to solve complicated problem and increasing the performance of the
model it is based on the concept of learning.
• Random forest is used for both classification and regression problems is the huge advantage.

7. Conclusion
• Evaluating the algorithmic rule, a dataset is collected, pre-processed, performed data modelling and studied a
value difference for the number of restricted days by the passengers for travelling.
• The results obtained by the random forest has better accuracy.
• In Upcoming days when huge amount of information is accessed as in detailed information in the dataset, the
expected results in future are highly correct.
• At last, it is curious to match our model accuracy with that of the business models accuracy offered
nowadays.

8. References
• Python for Every body - Dr. Charles Russell Severance
• Fundamentals of Machine Learning - John D. Kelleher