Customized Travel Route Recommendation on Big Social Media

ABSTRACT:

Big data increasingly benefit both research and industrial area such as health care, finance service and commercial recommendation. This paper presents a personalized travel sequence recommendation from both travelogues and community-contributed photos and the heterogeneous metadata (e.g., tags, geo-location, and date taken) associated with these photos. Unlike most existing travel recommendation approaches, our approach is not only personalized to user’s travel interest but also able to recommend a travel sequence rather than individual Points of Interest (POIs). Topical package space including representative tags, the distributions of cost, visiting time and visiting season of each topic, is mined to bridge the vocabulary gap between user travel preference and travel routes. We take advantage of the complementary of two kinds of social media: travelogue and community-contributed photos. We map both user’s and routes’ textual descriptions to the topical package space to get user topical package model and route topical package model (i.e., topical interest, cost, time and season). To recommend personalized POI sequence, first, famous routes are ranked according to the similarity between user package and route package. Then top ranked routes are further optimized by social similar users’ travel records. Representative images with viewpoint and seasonal diversity of POIs are shown to offer a more comprehensive impression. We evaluate our recommendation system on a collection of 7 million Flickr images uploaded by 7,387 users and 24,008 travelogues covering 864 travel POIs in 9 famous cities, and show its effectiveness. We also contribute a new dataset with more than 200K photos with heterogeneous metadata in 9 famous cities.

PROJECT OUTPUT VIDEO:

EXISTING SYSTEM:

• Existing studies on travel recommendation mining famous travel POIs and routes are mainly from four kinds of big social media, GPS trajectory, check-in data, geo-tags and blogs (travelogues). However, general travel route planning cannot well meet users’ personal requirements.

• Personalized travel recommendation recommends the POIs and routes by mining user’s travel records. The most famous method is location-based collaborative filtering (LCF). To LCF, similar social users are measured based on the location co-occurence of previously visited POIs. Then POIs are ranked based on similar users’ visiting records.

DISADVANTAGES OF EXISTING SYSTEM:

• Existing studies haven’t well solved the two challenges. For the first challenge, most of the travel recommendation works only focused on user topical interest mining but without considering other attributes like consumption capability.

• For the second challenge, existing studies focused more on famous route mining but without automatically mining user travel interest. It still remains a challenge for most existing works to provide both “personalized” and “sequential” travel package recommendation.

PROPOSED SYSTEM:

• We propose a Topical Package Model (TPM) learning method to automatically mine user travel interest from two social media, community-contributed photos and travelogues.

• To address the existing first challenge, we consider not only user’s topical interest but also the consumption capability and preference of visiting time and season. As it is difficult to directly measure the similarity between user and route, we build a topical package space, and map both user’s and route’s textual descriptions to the topical package space to get user topical package model (user package) and route topical package model (route package) under topical package space.

• Online module focuses on mining user package and recommending personalized POI sequence based on user package. First, tags of user’s photo set are mapped to topical package space to get user’s topical interest distribution. It is difficult to get user’s consumption capability directly from the textual descriptions of photos. But the topics user interested in could somehow reflect these attributes. For example, if a user usually takes part in luxurious activities like Golf and Spas, he is more likely to be rich.

• We combine user topical interest and the cost, time, season distribution of each topic to mine user’s consumption capability, preferred visiting time and season. After user package mining, we rank famous routes through measuring user package and routes package.

• At last, we optimize the top ranked routes through social similar users’ travel records in this city. Social similar users are measured by the similarity of user packages.

ADVANTAGES OF PROPOSED SYSTEM:

• Our work is a personalized travel recommendation rather than a general recommendation. We automatically mine user’s travel interest from user contributed photo collections including consumption capability, preferred time and season which is important to route planning and difficult to get directly.

• We recommend personalized POI sequence rather than individual travel POIs. Famous routes are ranked according to the similarity between user package and route package, and top ranked famous routes are further optimized according to social similar users’ travel records.

• We propose Topical Package Model (TPM) method to learn user’s and route’s travel attributes. It bridges the gap of user interest and routes attributes. We take advantage of the complementary of two big social media to construct topical package space.

MODULES:

• Social Media Mining System Construction

• User Topical Package Model Mining

• Route Package Mining

• Travel sequence recommendation

MODULES DESCSRIPTION:

Social Media Mining System Construction

• In the first module we develop the system for the evaluation of our proposed model and thus make the system construction module with social media mining system.

• Our topic package space is the extension of textual descriptions of topics such as ODP. We use the topical package space to measure the similarity of the user topical model package (user package) and the route topical model package (route package). In our paper, we construct the topical package space by the combination of two social media: travelogues and community-contribute photos. To construct topical package space, travelogues are used to mine representative tags, distribution of cost and visiting time of each topic, while community-contributed photos are used to mine distribution of visiting time of each topic.

• The reasons for using the combination of social media are (1) travelogues are more comprehensive to describe a location than the tags with the photos which are with so many noises; (2) it is difficult to mine a user’s consumption capability and the cost of POIs directly by the photos or the tags with the photos; (3) to season, although both media could offer correct visiting season information of POIs, the number of photos of a POI is far larger than the number of travelogues. (4) the time difference between where the user lives and the “data taken” of community contributed photos of where he or she visits make the taken time inaccurate.

User Topical Package Model Mining

• User topical package model (user package) is learnt from mapping the tags of user’s photos to topical package space. It contains user topical interest distribution
  (U), user consumption capability (U), preferred travel time distribution (U) and preferred travel season distribution .

• In this module, we introduce how to extract the user package, which contains user topical interest distribution, user consumption capability distribution, preferred travel time distribution and preferred travel season distribution.

• First we introduce user’s topical interest mining from mapping user’s tags to the topical package space. Then, we introduce how to get topical space mapping method.

• We map the textual description (tags) of user‘s community photos to the topical package space to present the user’s travel preference of different topics, which is defined as user topical interest distribution. We assume that if a user’s tags appear frequently in one topic and less in others, the user has a higher interest towards this topic.

• We use the cost distributions of the all the topics and distribution of use’s topical interest to present a user’s consumption capability. If a user usually takes part in luxurious activities like Golf and Spas, his consumption capability is very likely to be. If a user usually takes part in some cheap things, his consumption capability is likely to be low, and we tend not to recommend him luxurious topics.

Route Package Mining

• Route topical package model (route package) is learnt from mapping the travelogues related to the POIs on the route to topical package space. It contains route topical interest, route’s cost distribution, route’s time distribution and season distribution.

• To save the online computing time, we mine travel routes and the attribute of the routes offline. After mining POIs, to construct travel routes, we analyze the spatio-temporal structure of the POIs among travelers’ records.

• We construct the spatio-temporal structure of the POIs according to the “data taken”. POI with the earlier timestamp is defined as the “in”. POI with a later timestamp, on the contrary, is defined as “out”. Then we count the times of “in” and “out” from POI to others by the records of all the users after filtering. A greedy algorithm is then applied to find the time sequence of these POIs. Thus, we finish famous routes mining and obtain famous routes of each city.

Travel sequence recommendation

• After mining user package and route package, in this module, we develop our travel routes recommendation module. It contains two main steps: (1) routes ranking according to the similarity between user package and routes packages, and (2) route optimizing according to similar social users’ records.

• After POI and route ranking module, we get a set of ranked routes. Here, we further describe the optimization of top ranked routes according to social similar users’ travel records. Firstly, we introduce how to mine social similar users and their travel records. Then we introduce how to optimize the roads by social users’ travel records.

SYSTEM REQUIREMENTS:

HARDWARE REQUIREMENTS:

• System           : Pentium Dual Core.

• Hard Disk      : 120 GB.

• Monitor          : 15’’ LED

• Input Devices : Keyboard, Mouse

• Ram                : 1GB.

SOFTWARE REQUIREMENTS:

• Operating system : Windows 7.

• Coding Language : ASP.NET,C#.NET

• Tool                       : Visual Studio 2008

• Database                : SQL SERVER 2005