TRAVEL PACKAGE RECOMMENDATION BY USING USER PREFERENCES

TRAVEL PACKAGE RECOMMENDATION BY USING USER PREFERENCES

Archana.R

, Sankar Ganesh.K

1

Student ME/CSE, Kingston engineering college, (India)

2

Assistant professor CSE, Kingston engineering college, (India)

ABSTRACT

There are many recommendation system for tour packages, but they fail to create a package that suite the customer need. The unique characteristics of travel data are area of interest, session and travel mode. The tour spots are distributed in many geographical locations. The topic extraction is conditioned on both the tourists and the intrinsic features like locations, travel seasons, mode of transport. The TRAST model scans the locations where most of the users like. Then the locations are filtered by session (eg: winter, summer). Then the filtered set is again filtered by travel mode like bus, car, van etc… Finally a tour package is created which is the best suited to the customer needs

Keywords: Recommendation System, Tour

I. INTRODUCTION

There are many online travel companies in this world. Each day there are new travel companies emerging. There is a confusion for the tourists to choose which travel company to travel. Also in a travel company, there are many packages available. Most of the time, travelers choose a package which contains a set of travel spots. Not all the spots are liked by the user, the user likes only one or few of the package. This paper develops a recommendation system which can be used by Travel Company to create a package where the package contains a list of tour spots that are fully liked by the travelers.

The recommendation system is applied in many fields [1] [2] [3]. There are many problems in recommending a travel package. The first one is there is less amount of data as while comparing to other recommendation system like movie recommendation. The second problem is each travel spot contain complex relationship like landscapes, area, session etc…

Tour packages are dynamic that is the new packages will be replaced by old packages over time to attract more customers.

There are two types of packages, static and dynamic. Static packages do not change over time. But dynamic packages change over time.

Our paper shows three algorithm to develop an efficient dynamic package. A new package is always a cold-start problem because the packages have a life time and it will be only active in that particular life time. This problem can be overcome by creating a package that is similar to already existing one. The new package should also include all the current trends of the user.

II. ARCHITECTURE DIAGRAM

Figure 1 Architecture Diagram

There are many tour spots, each user picks the tour spots of his interest, then the maximum chosen spots are collected and a package is created.

III. TAST MODEL (TOURISTS AREA SESSION TOPIC)

This model represents different landscapes and tourists based on topic distributions. In the TAST model, the extraction of topics is conditioned on both the tourists and the intrinsic features (i.e., locations, travel seasons) of the landscapes. As a result, the TAST model can well represent the content of the travel packages and the interests of the tourists. Based on this TAST model, a cocktail approach is developed for personalized travel package recommendation by considering some additional factors including the seasonal behaviors of tourists, the prices of travel packages, and the cold start problem of new packages. Finally, the experimental results on real-world travel data show that the TAST model can effectively capture the unique characteristics of travel data and the cocktail recommendation approach performs much better than traditional techniques.

3.1 Algorithm

Step 1: open the database connection Step2: create a linked list

Step3: search for the areas with the string “select * from area where areaname=’L’”

Step4: for each result in the select query Step5: add the result to linked list Step6: return the linked list

IV TRAST MODEL (TOURIST RELATION AREA SESSION TOPIC)

TRAST helps to understand the reasons why tourists form a travel group. This goes beyond personalized package recommendations and is helpful for capturing the latent relationships among the tourists in each travel group. In addition, we conduct systematic experiments on the realworld data. These experiments not only demonstrate that the TRAST model can be used as an assessment for travel group automatic formation but also provide more insights into the TAST model and the cocktail recommendation approach. In summary, the contributions of the TAST model, the cocktail approaches, and the TRAST model for travel package recommendations are shown in Fig. 1, where each dashed rectangular box in the dashed circle identifies a travel group and the tourists in the same travel group are represented by the same icons.

4.1 Algorithm

Input: request for a package Output: a recommended package Algorithm:

Step1: monitor the user request for a given time Step2: get the most popular areas

Step3: list the areas

Step4: the admin selects few area from the list Step5: create a new package

Step6: admin enters the prince Step7: return the package

A tour package is comprises transport and accommodation advertised and sold together by a vendor known as a tour operator. Other services may be provided such a rental car, activities or outings during the holiday.

Transport can be via charter airline to a foreign country, and may also include travel between areas as part of the holiday. Package holidays are a form of product bundling.

Figure 2: An Example of Tour Package

Dynamic packaging is a method used in package holiday bookings to enable consumers to build their own package of flights, accommodation, and car rental instead of purchasing a pre-defined package.[7] Dynamic packages differ from traditional package tours in that the pricing is always based on current availability,

escorted group tours are rarely included, and trip-specific add-ons such as airport parking and show tickets are often available.

Dynamic packages are similar in that often the air, hotel, and car rates are available only as part of a package or only from a specific seller. The term "dynamic packaging" is often used incorrectly to describe the less sophisticated process of interchanging various travel components within a package, however, this practice is more accurately described as "dynamic bundling".

True dynamic packaging demands the automated recombination of travel components based on the inclusion of rules that not only dictate the content of the package but also conditional pricing rules based on various conditions such as the trip characteristics, suppliers contributing components, the channel of distribution, and terms of sale.[8]

Dynamic packages are primarily sold online, but online travel agencies will also sell by phone owing to the strong margins and high sale price of the product.

V AREA SESSION SEGMENTATION

Tourist area are segmented as north India, east India, south India and west India. The sessions are segmented as summer, winter, rainy.

All the travel logs are saved in the database. The collaborative filtering is used to retrieve the maximum travel spots that are used by the users. Ten the inactive spots are removed and a new package is recommended.

Figure 3, The Cocktail Approach 5.1 Create Package – Algorithm

Input: package name

Output: list of areas in the package Algorithm:

Step1: get the package name

Step2: create a package using a insert query Step3: add a list of areas in the package Step4: a area contain name, price, session, type Step5: repeat 4 for all the area names

REFERENCES

[1] G. Adomavicius and A. Tuzhilin, “Toward the Next Generation of Recommender Systems: A Survey of the State-of-the-Art and Possible Extensions,” IEEE Trans. Knowledge and Data Eng., vol. 17, no. 6, pp.

734-749, June 2005.

[2] Y. Ge et al., “An Energy-Efficient Mobile Recommender System,” Proc. 16th ACM SIGKDD Int’l Conf.

Knowledge Discovery and Data Mining (SIGKDD ’10), pp. 899-908, 2010

[3] B. Sarwar, G. Karypis, J. Konstan, and J. Riedl, “Application of Dimensionality Reduction in Recommender Systems-a Case Study,” Proc. ACM WebKDD Workshop, pp. 82-90, 2000.

[4] M. Xie, L.V.S. Lakshmanan, and P.T. Wood, “Breaking Out of the Box of Recommendations: From Items to Packages,” Proc. Fourth ACM Conf. Recommender Systems (RecSys ’10), pp. 151-158, 2010.

[5] Y. Ge et al., “Cost-Aware Travel Tour Recommendation,” Proc. 17th ACM SIGKDD Int’l Conf.

Knowledge Discovery and Data Mining (SIGKDD ’11), pp. 983-991, 2011.

[6] http://en.wikipedia.org/wiki/Package_tour

[7] "Travel Agents Could Lose Out in the Dynamic Packaging Battle". First Conferences Ltd. 2005. Archived from the original on 6 February 2005. Retrieved 19 January 2005.

[8] Method and apparatus for the composition and sale of travel-oriented packages". United States Patent Number 7,136,821 Inventors: Kohavi, Bar-david, et al. 2006. Retrieved 16 November 2006.