Automatic Identification System based Fishing Trajectory Data Preprocessing Method using Map Reduce
Jin-wan Park1, Keon Myung Lee2, Kwang-il Kim3
1Jin-wan Park, Graduate, Department of Maritime Transportation System, Mokpo National Maritime University, Mokpo, Republic of Korea.
2Keon Myung Lee, Department of Computer Science, Chungbuk National University, 1 Chungdae-ro Cheongju-si, Republic of Korea.
3Kwang-il Kim, Division of Marine Industry and Maritime Police, Jeju National University, Jeju, Republic of Korea.
Manuscript received on 19 August 2019 | Revised Manuscript received on 29 August 2019 | Manuscript Published on 16 September 2019 | PP: 352-356 | Volume-8 Issue-2S6 July 2019 | Retrieval Number: B10670782S619/2019©BEIESP | DOI: 10.35940/ijrte.B1067.0782S619
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Many countries use vessel monitoring system (VMS) data to monitor their fishery activities. However, VMS data is limited in terms of distinguishing operations involving illegal fishing gear. Recently introduced automatic identification system (AIS) data is advantageous for tracking fishing ship behaviors.AIS data include various types of information about a ship, such as its state of navigation and its broadcast rate on the radio channel. We interpolate AIS trajectory data with a regular time interval and extract the ship velocity and course change data for fishing ship gear classification. To simplify and condense the data, the course change index (CCI) and ship speed index (SSI) are applied to the ship velocity and course data. The proposed mapper combines CCIs and SSIs into key words, while the proposed reducer collects fishing ship gear type values that are of the same key.By using the proposed key-value dataset from the MapReduce procedure, we can classify fishing gear type. We evaluated the performance of the proposed model by using a test dataset. The results showed that the proposed model achieved 76.2% accuracy in the classification of fishing ship trajectories against the test dataset.
Keywords: Automatic Identification System, Course Change Index, Fishing Activity, Fishing Gear Classification, MapReduce, Vessel Monitoring System.
Scope of the Article: Data Analytics