A Data-Mining Approach to Travel Price Forecasting

https://hal.archives-ouvertes.fr/hal-00665041/document

https://hal.archives-ouvertes.fr/hal-00665041/document
A Data-Mining Approach to Travel Price Forecasting


http://www.cs.ru.nl/bachelorscripties/2014/Tim_Janssen___4150880___A_Linear_Quantile_Mixed_Regression_Model_for_Prediction_of_Airline_Ticket_Prices.pdf
A Linear Quantile Mixed Regression Model for Prediction of Airline Ticket Prices 


http://www.isi.edu/info-agents/papers/etzioni03-kdd.pdf


To Buy or Not to Buy: Mining Airfare Data to Minimize Ticket Purchase Price Oren Etzioni Dept. Computer Science University of Washington Seattle, Washington 98195 etzioni@cs.washington.edu Craig A. Knoblock Information Sciences Institute University of Southern California Marina del Rey, CA 90292 knoblock@isi.edu Rattapoom Tuchinda Dept. of Computer Science University of Southern California Los Angeles, CA 90089 pipet@isi.edu Alexander Yates Dept. Computer Science University of Washington Seattle, Washington 98195 ayates@cs.washington.edu ABSTRACT As product prices become increasingly available on the World Wide Web, consumers attempt to understand how corporations vary these prices over time. However, corporations change prices based on proprietary algorithms and hidden variables (e.g., the number of unsold seats on a flight). Is it possible to develop data mining techniques that will enable consumers to predict price changes under these conditions? This paper reports on a pilot study in the domain of airline ticket prices where we recorded over 12,000 price observations over a 41 day period. When trained on this data, Hamlet — our multi-strategy data mining algorithm — generated a predictive model that saved 341 simulated passengers $198,074 by advising them when to buy and when to postpone ticket purchases. Remarkably, a clairvoyant algorithm with complete knowledge of future prices could save at most $320,572 in our simulation, thus Hamlet’s savings were 61.8% of optimal. The algorithm’s savings of $198,074 represents an average savings of 23.8% for the 341 passengers for whom savings are possible. Overall, Hamlet saved 4.4% of the ticket price averaged over the entire set of 4,488 simulated passengers. Our pilot study suggests that mining of price data available over the web has the potential 



http://community.loyola.edu/dchillman/wp-content/uploads/sites/389/2016/03/Case-Assignment-Part-A.pdf 

airfairs 

Danielle Hillman Prof. Paul Lande EC 220 03 18 February 2016


http://www.graphpad.com/guides/prism/6/statistics/index.htm?stat_skewness_and_kurtosis.htm
http://cs229.stanford.edu/proj2012/Papadakis-PredictingAirfarePrices.pdf

Predicting Airfare Prices Manolis Papadakis Introduction Airlines implement dynamic pricing for their tickets, and base their pricing decisions on demand estimation models. The reason for such a complicated system is that each flight only has a set number of seats to sell, so airlines have to regulate demand. In the case where demand is expected to exceed capacity, the airline may increase prices, to decrease the rate at which seats fill. On the other hand, a seat that goes unsold represents a loss of revenue, and selling that seat for any price above the service cost for a single passenger would have been a more preferable scenario. The purpose of this project was to study how airline ticket prices change over time, extract the factors that influence these fluctuations, and describe how they're correlated (essentially guess the models that air carriers use to price their tickets). Then, using that information, build a system that can help consumers make purchasing decisions by predicting how air ticket prices will evolve in the future. We focused our efforts on coach-class fares. Related Work There has been some previous work on building prediction models for airfare prices using Machine Learning techniques [1] [2] [3]. The various research groups have focused on mostly different sets of features and trained their models on different kinds of flights. A major distinction among these projects is the specific trend they are trying to predict. Specifically, we can categorize projects into 2 approaches: studying the factors that influence the average price of a flight [2], or those that influence the price of a specific flight in the days leading up to departure [1] [3]. We will use this distinction in the definition of our model. There also exist commercial services, like Bing Travel (which actually evolved from the work in [1]), that perform this kind of prediction, but their models are not made public. 


Quantum Field Theory II An introduction to Feynman diagrams A course for MPAGS Dr Sam T Carr University of Birmingham carrst@theory.bham.ac.uk February 9, 2009


Linear response theory and the fluctuation-dissipation theorem There is a very strong link between correlation functions (fluctuations) and response functions to an external applied field (dissipation). For example, consider the expectation value %GS|ρˆ(!x,t)ρˆ(!x " ,t " )|GS#. (8.1) We have been thinking of this as a correlation function between two points and times. However, we can also think of it as applying some extra density at !x " ,t " and seeing how it affects the density at x,t. Time ordered correlation functions are what we have been learning to calculate using the diagrammatic technique. However, response functions are what experimentalists will measure - they will perturb the system in some way and see what it does in response. It is therefore very important for us to understand much better the relationship between response functions and correlation functions. In fact, the fluctuation-dissipation theory in it’s full glory can not be fully understood until we have looked at finite temperature Green functions - so we will revisit many of these concepts again later in Lecture 12. However, the importance of this topic means it is well worthwhile meeting it at least once before we leave the subject of zero temperature Green functions - which is the purpose of this lecture.


http://www.physics.miami.edu/~nearing/mathmethods/operators.pdf

Operators and Matrices You’ve been using operators for years even if you’ve never heard the term. Differentiation falls into this category; so does rotation; so does wheel-alignment. In the subject of quantum mechanics, familiar ideas such as energy and momentum will be represented by operators. You probably think that pressure is simply a scalar, but no. It’s an operator. 

https://www.math.stonybrook.edu/~scott/mat308.spr11/TraceDet.pdf

Linearization, Trace and Determinant MAT308 Spring 2011 Scott Sutherland, Stony Brook University In order to analyze a system of nonlinear equations, one important aspect is the idea of linearization near a fixed point (or equilibrium solution). More specifically, suppose that we have a system of the form dX~ dt = F~ (X~ ), and furthermore that for some initial condition X~ 0 we have F~ (X~ 0) = ~0. Then the constant solution X~ (t) = X~ 0 is an equilibrium. That is, the trajectory in the phase plane is just a single point. We want to analyze what happens to solutions that start near such a solution

https://www.cs.umn.edu/sites/cs.umn.edu/files/tech_reports/11-025.pdf

 A regression model for predicting optimal purchase timing for airline tickets William Groves and Maria Gini Department of Computer Science and Engineering, University of Minnesota {groves, gini}@cs.umn.edu Abstract Optimal timing for airline ticket purchasing from the consumer’s perspective is challenging principally because buyers have insufficient information for reasoning about future price movements. This paper presents a model for computing expected future prices and reasoning about the risk of price changes. The proposed model is used to predict the future expected minimum price of all available flights on specific routes and dates based on a corpus of historical price quotes. Also, we apply our model to predict prices of flights with specific desirable properties such as flights from a specific airline, non-stop only flights, or multi-segment flights. By comparing models with different target properties, buyers can determine the likely cost of their preferences. We present the expected costs of various preferences for two high-volume routes. Performance of the prediction models presented is achieved by including instances of time-delayed features, by imposing a class hierarchy among the raw features based on feature similarity, and by pruning the classes of features used in prediction based on in-situ performance. Our results show that purchase policy guidance using these models can lower the average cost of purchases in the 2 month period prior to a desired departure. The proposed method compares favorably with a deployed commercial web site providing similar purchase policy recommendations. 1 Introduction Adversarial risk in the airline ticket domain exists in two contexts: the adversarial relationship between buyers and sellers, and the competitive relationships that exist between multiple airlines providing the equivalent service. Buyers are often seeking the lowest price on their travel, while sellers are seeking to keep overall revenue as high as possible to maximize profit. Simultaneously, each seller must consider the price movements of its competitors to ensure that its prices remain sufficiently competitive to achieve sufficient (but not too high) demand. It is impossible to effectively address the problem of optimizing decision making from the buyer’s point of view without also considering both types of adversarial relationships. Sellers (airlines) make significant long term investments in fixed infrastructure (airports, repair facilities), hardware (planes), and route contracts. The specific details of these long term decisions are intended to roughly match expected demand but often do not match exactly. Dynamic setting of prices is the mechanism that airlines use to increase the matching between their individual supply and demand profile in order to attain the greatest revenue. A central challenge in the airline ticket purchasing domain is the information asymmetry that exists between buyers and sellers. Airlines have the ability to mine significant databases of historical sales data to develop models for expected future demand for each flight. Demand for a specific flight is likely to vary over time and will also vary based on the pricing strategy adopted by the airline. For buyers, it is generally best to buy far in advance of a flight’s departure because the prices tend to increase dramatically as the departure date approaches. But, airlines often violate this principle and adjust prices downward to increase sales. We make two novel contributions in this work: (1) a method of automated optimal feature set generation from the data that leverages a hierarchicalization of the available features to efficiently compute a feature set is proposed; (2) the addition of time-delayed observations to the feature vector fed to the machine learning 1 algorithm is performed. This allows anticipation of trends and more complex relationships between variables. For instance, we address pricing behaviors up to and beyond 60 days prior to departure, and we consider purchasing a flight on any airline for a specific date and city pair (previous work only considers the cost of a specific pair of flight numbers from two specific airlines). These ideas are then experimentally applied to prediction in the real-world airline ticket purchase domain. This paper presents models that also accommodate preferences of passengers about the number of stops in the itinerary or the specific airline to use. We believe this prediction task is both a more difficult task and generates models that are more useful for actual airline passengers.

http://liawww.epfl.ch/uploads/project_descriptions/project_344.pdf

Machine Learning for Predicting Flight Ticket Prices Supervisor: Igor Kulev November 9, 2015 1 Overview and Goal Airline companies have the freedom to change the flight ticket prices at any moment. Travellers can save money if they choose to buy a ticket when its price is the lowest. The problem is how to determine when is the best time to buy flight ticket for the desired destination and period. Airline companies use many different variables to determine the flight ticket prices: indicator whether the travel is during the holidays, the number of free seats in the plane etc. Some of the variables are observed, but some of them are hidden. The goal of this project is to use machine learning techniques to model the behavior of flight ticket prices over the time. In order to build and evaluate the model, you will use data that contains historical flight ticket prices for particular routes. 2 Project Steps • Parse and preprocess the flight ticket data that will be given to you • Apply few different machine learning models on the data set • Evaluate and compare the models 3 Required Skills The student should have knowledge in Machine learning. The student should know how to work with some programming language for machine learning (for example Matlab). References [1] O. Etzioni, R. Tuchinda, C. A. Knoblock, and A. Yates, “To buy or not to buy: mining airfare data to minimize ticket purchase price,” in Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining. ACM, 2003, pp. 119– 128