Presentation 2021-11-26
[Poster Presentation] Fish Catch Forecasting by Species using Smart Buoy Sensing Data
Cong Cao, utsunomiya eiji, yoshihara kiyohito,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Fish catch forecasting based on IoT(Internet of Things) data is an important theme for smart fishery, which is expected to solve the problems like know-how humanize and insufficient workers. In this paper, we build regression machine learning models (ResNet Regressor and Random Forest Regressor) based on sensor data and weather data to forecast fish catch weight from a local fishing ground in Japan. From the analysis of the relevance between features and weight target, we find that when compared with the total weight, weight target for each fish spices is more related to environment features like water temperature and atmospheric temperature. Otherwise, lack of enough data to train the model is also a big problem because the harsh environments often cause IoT system sensing failures. To solve the problem, 2 methods are discussed in this paper. The first method, train machine learning models to predict sensor values based on weather data, and use the prediction as missing data imputation. The second method, use data from several different fishing grounds. We conduct experiments to evaluated the above 2 methods for total weight and weights for some fish species. As a result, method 1 is effective for all fish species, especially effective for [Jack mackerel], which reduce the RMSE of ResNet Regressor form 22.12 to 18.26. Method 2 is effective for [Jack mackerel] [Silver-stripe round herring] and [Unicorn leatherjacket], also especially effective for [Jack mackerel], which reduce the RMSE of Random Forest Regressor form 29.27 to 26.09.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Fish catch forecasting / Weight target for each fish spices / Machine learning
Paper # SRW2021-46,SeMI2021-45,CNR2021-20
Date of Issue 2021-11-18 (SRW, SeMI, CNR)

Conference Information
Committee SRW / SeMI / CNR
Conference Date 2021/11/25(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Kikai-Shinko-Kaikan Bldg.
Topics (in Japanese) (See Japanese page)
Topics (in English) IoT Workshop
Chair Hanako Noda(Anritsu) / Koji Yamamoto(Kyoto Univ.) / Masayuki Kanbara(NAIST)
Vice Chair Keiichi Mizutani(Kyoto Univ.) / Kentaro Saito(Tokyo Denki Univ.) / Hirokazu Sawada(NICT) / Kazuya Monden(Hitachi) / Yasunori Owada(NICT) / Yoshihiko Murakawa(Kyoto Inst. of Tech.)
Secretary Keiichi Mizutani(NTT) / Kentaro Saito(NIigata Univ.) / Hirokazu Sawada(Cyber Univ.) / Kazuya Monden(Waseda Univ.) / Yasunori Owada(Osaka Univ.) / Yoshihiko Murakawa(Toshiba)
Assistant Akihito Noda(Nanzan Univ.) / Yuki Katsumata(NTT DOCOMO) / Akihito Taya(Aoyama Gakuin Univ.) / Yu Nakayama(Tokyo Univ. of Agri. and Tech.) / Yuta Hoshi(NHK) / Junji Yamato(Kogakuin Univ.)

Paper Information
Registration To Technical Committee on Short Range Wireless Communications / Technical Committee on Sensor Network and Mobile Intelligence / Technical Committee on Cloud Network Robotics
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) [Poster Presentation] Fish Catch Forecasting by Species using Smart Buoy Sensing Data
Sub Title (in English)
Keyword(1) Fish catch forecasting
Keyword(2) Weight target for each fish spices
Keyword(3) Machine learning
1st Author's Name Cong Cao
1st Author's Affiliation KDDI Research, Inc(KDDI Research)
2nd Author's Name utsunomiya eiji
2nd Author's Affiliation KDDI Research, Inc(KDDI Research)
3rd Author's Name yoshihara kiyohito
3rd Author's Affiliation KDDI Research, Inc(KDDI Research)
Date 2021-11-26
Paper # SRW2021-46,SeMI2021-45,CNR2021-20
Volume (vol) vol.121
Number (no) SRW-265,SeMI-266,CNR-267
Page pp.pp.65-67(SRW), pp.52-54(SeMI), pp.42-44(CNR),
#Pages 3
Date of Issue 2021-11-18 (SRW, SeMI, CNR)