Design And Implementation Of Personalized Trip Route Search System

With the rapid economic development and the people's standard of living has risen steadily, more and more people like to patronize travel for relaxing and entertainment. With the rapid development and popularization of computer and internet technology in recent years, the amount of internet users increased rapidly, and e-commerce has gradually occupied the market. Consequently, more and more people popular to online shopping, which provides a convenient and efficient platform for both enterprises and users. With no doubt, the number of people who choose to search the Internet for better tourist routes keeps increasing. However, various tourism web sites show their own distinguished, such as Way Cattle Travel, Showtime Travel Sites, Travel Ant Cellular Network. The information and price provided by tourism web sites vary greatly; accordingly, people always need to search plenty of websites to get quite a few plans if they want a low-cost tour; and then they themselves have to compare the price and all other factors before they may get a preferable result.In allusion to the problems mentioned above, the thesis presents a personalized trip route system, which is proposed to obtain travel information provided by a number of travel sites, and then aggregate and compare this information, as well as analyze the relevant routes. Consequently, the system paves a new way for users to make the best choice.We first crawl all travel information from different travel websites via Web Crawler-a frame based on Python, and the web pages are stored in a distributed system in the meantime. Then from the stored pages, we can extract key information including product name, departure city, destination, price, departure time, means of transportation, tourist attractions, hotels, travel arrangements, etc. The next step is to configure the task scheduler subsystem to control the frequency of reptiles, the start time and the end time, etc. If a user types in a departure city, a destination city, the departure time and other factors, then based on some certain logic, the system will display a comparison result and several routes to help the user to make decisions. The sequencing module has been plugged into this system. We can cluster the users* information with k-means algorithm, analyze users'consuming habits, and then adjust the order of tourist routs. As a result, if a rout is chosen by most users, it will be displayed in the first place, which will shorten users'time for searching and choosing.The proposed system is implemented to facilitate user's search. Consequently, if some tourism routes are highly focused, they will be of high priority when recommended to users by this system.