Data Mining from Crawled data in a Hadoop system.

Data Mining Project.(Prototype) 1. Overview - This is Infowork inc. in Korea. ([login to view URL]) - We are try to be a new frontier of the Big Data market in Korea. - The main goal of this project is to find the partners who can work with us by long term based. - For the first step, we are checking the possibility of this offshore collaboration. - The raw data will be prepared by us before you start this project. - The detail mining requirements will be defined on this Saturday(1st / June), before 8pm (GMT+9). : hope to get some proposal from you (new idea or approach). : If the proposal is great and reasonable, the total project budget can be upgraded. 2. Required Skills : - Statistics, NLP(Natural Language Processing), Data Mining, SQL 3. Key intention for the mining(background of this project) : - Our client wants to know what makes travelers to be satisfied or happy when they come to a place for tourism. - Every travelers have their own purposes and points of view for the specific travels. (Business , Leisure, honeymoon, sightseeing, etc.) - But, our client want to find some clues how to increase the number of the tourists in specific spot (country, place, event, etc.) 4. Key Requirements : - Proposal of diverse points of view for the analysis to find the clue mentioned above. - Report by BIRT (need to be agreed with us) - All the outputs(formula and code and SQL and BIRT setting and etc.) which are used and created for this project must be submitted before the last milestone released. - Additional detail requirements can be defined in the progress of this project. (Of course by communication) 5. Reference : - The raw data are extracted by the keywords : 'Korea' and ('travel*' or 'trip' or 'tour' or 'journey') - Target Data : crawled twit logs for 7 days + crawled data of 'trip [login to view URL]' (duration not fixed yet) ** Twit data can be increase by some needs. - Total amount of target Data : is Not fixed yet (currently in processing) 6. Method of payment - Three milestone will be issued : + First milestone : Issued on the start date by 10% of the total budget / Released by fixing the detail requirements. + Second milestone : Issued right after the first release by 40% of the total budget / Released by the completion of the reports agreed. + Third milestone : Issued right after the second release by 50% of the total budget / Released by the completion of the verification for the submitted documents. Duration of the project : In 5 days (Negotiable)