Shuhei Goda

ウォンテッドリー株式会社 / データサイエンティスト

Shuhei Goda

ウォンテッドリー株式会社 / データサイエンティスト

Shuhei Goda

ウォンテッドリー株式会社 / データサイエンティスト

Why not create your own Wantedly profile?

Showcase yourself and your experiences to bring your professional story to life.

Join

Connect to learn more

You'll be able to see their introduction and other information once they have accepted your connection request.

Kaggle Google QUEST Q&A Labeling: 23th / 1571 (silver)

atma杯 オンサイトデータコンペ #2 5th / 39

Kaggle Freesound Audio Tagging 2019 Competition: 23th / 880 (silver)

Kaggle Instant Gratification Competition: 64th / 1836 (silver)

Kaggle Santander Value Prediction Challenge Competition: 62th / 4484 (silver)

Kaggle PLAsTiCC Astronomical Classification Competition: 16th / 1094 (silver)

In the future

Ambition

In the future

Visitに関わる全ての人により良い体験をして頂けるよう、プロダクトに対して効果的な施策を立案・実装し続けていきたいです。 また、モデリングスキルだけでなく、より広範囲のエンジニアリングスキルを身につけていき、自分の取り組める仕事の範囲を広げていきたいと思っています。

Sept 2019
-
Present

データサイエンティスト
Present

Sept 2019 -

Present

検索・推薦システムの構築・改善を行っています

DEIM2021技術報告 / 会社訪問アプリ「Wantedly Visit」のデータで見る相互推薦システム

Mar 2021

DEIM2021技術報告 / 会社訪問アプリ「Wantedly Visit」のデータで見る相互推薦システム

Mar 2021

Weighted Averaging of Various LSTM Models for Next Destination Recommendation

This paper describes the 6th place approach to Booking.com WSDM WebTour 2021 Challenge, which is a challenge with a task of predicting travellers’ next destination. We, in the team "hakubishin3 & u++ & yu-y4", trained four types of Long short-term memory (LSTM) models, and achieved the final score: 0.5399 by weighted averaging of these predictions. There are some differences in these models in feature engineering, multi-task learning, and data augmentation. Our experiments showed that the diversity of the models boosted the final result. Our codes are available at https://github. com/hakubishin3/booking-challenge-2021 and https://github.com/ upura/booking-challenge-2021.

Mar 2021

Weighted Averaging of Various LSTM Models for Next Destination Recommendation

Mar 2021

Wantedly RecSys 2020 参加レポート⑥ - Offline Evaluation in Recommender Systems

Oct 2020

Wantedly RecSys 2020 参加レポート⑥ - Offline Evaluation in Recommender Systems

Oct 2020

RecSys Challenge 2020 Workshop: A Stacking Ensemble Model for Prediction of Multi-type Tweet Engagements

Sept 2020

RecSys Challenge 2020 Workshop: A Stacking Ensemble Model for Prediction of Multi-type Tweet Engagements

Sept 2020

A Stacking Ensemble Model for Prediction of Multi-type Tweet Engagements

The RecSys Challenge 2020 is a competition with a task of predicting four types of user engagements on Twitter: Like, Reply, Retweet and Retweet with comment. In this paper, we describe Team Wantedly’s approach to this challenge, which won the third place. We found that the targets are highly correlated and it is important to use every engagement to predict the other engagements. Therefore, we choose to stack LightGBM models to use this co-occurrences effectively in the large dataset. Our final scores are as follows: 1.5266 (Retweet PR-AUC), 30.06 (Retweet RCE), 0.1918 (Reply PR-AUC), 20.44 (Reply RCE), 0.7716 (Like PR-AUC), 24.76 (Like RCE), 0.0724 (Retweet with comment PR-AUC), 14.86 (Reply RCE). Our code is available at https://github.com/wantedly/recsys2020-challenge. Shuhei Goda, Naomichi Agata, and Yuya Matsumura. 2020. A Stacking Ensemble Model for Prediction of Multi-Type Tweet Engagements. In Proceedings of the Recommender Systems Challenge 2020. 6-10.

Sept 2020

A Stacking Ensemble Model for Prediction of Multi-type Tweet Engagements

Sept 2020

RecSys Challenge 2020: 3rd (prize winning)

June 2020

RecSys Challenge 2020: 3rd (prize winning)

June 2020

Kaggle Google Quest Q&A Labeling - 23rd place solution

Feb 2020

Kaggle Google Quest Q&A Labeling - 23rd place solution

Feb 2020

Kaggle Days Tokyo Report #2

Dec 2019

Kaggle Days Tokyo Report #2

Dec 2019

Kaggle Days Tokyo Competition: 5th / 88

Dec 2019

Kaggle Days Tokyo Competition: 5th / 88

Dec 2019

Target Encoding はなぜ有効なのか

Nov 2019

Target Encoding はなぜ有効なのか

Nov 2019

Kaggle Masterになるまでを振り返る

Oct 2019

Kaggle Masterになるまでを振り返る

Oct 2019

Kaggle IEEE-CIS Fraud Detection Competition: 5th / 6831(gold)

Oct 2019

Kaggle IEEE-CIS Fraud Detection Competition: 5th / 6831(gold)

Oct 2019

The Web Conference2020 参加報告会 - 論文・セッション紹介

Apr 2020

The Web Conference2020 参加報告会 - 論文・セッション紹介

Apr 2020

Apr 2016
-
Aug 2019

Apr 2016 - Aug 2019

統計・機械学習を使ってクライアントのビジネス課題の解決に取り組んできました。一部の案件では分析チームのリードを担当しました。

atma杯 オンサイトデータコンペ #1 2nd / 32

Aug 2019

atma杯 オンサイトデータコンペ #1 2nd / 32

Aug 2019

【論文紹介】EX2: exploration with exemplar models for deep reinforcement learning

Feb 2018

【論文紹介】EX2: exploration with exemplar models for deep reinforcement learning

Feb 2018

Mar 2016

北海道大学大学院理学院

宇宙理学専攻

Mar 2016

惑星観測用補償光学装置の開発

天体写真

天体写真

Mar 2014

北海道大学

理学部 地球惑星科学科

Mar 2014

惑星観測用補償光学装置の開発


Skills and qualities

kaggle

Recommended by 一條 端澄, Yamaguchi Takahiro
2

機械学習

Recommended by 一條 端澄, Yamaguchi Takahiro
2

Python

Recommended by Yamaguchi Takahiro
1

Publications

DEIM2021技術報告 / 会社訪問アプリ「Wantedly Visit」のデータで見る相互推薦システム

Mar 2021

Weighted Averaging of Various LSTM Models for Next Destination Recommendation

Mar 2021

Wantedly RecSys 2020 参加レポート⑥ - Offline Evaluation in Recommender Systems

Oct 2020

RecSys Challenge 2020 Workshop: A Stacking Ensemble Model for Prediction of Multi-type Tweet Engagements

Sept 2020

A Stacking Ensemble Model for Prediction of Multi-type Tweet Engagements

Sept 2020

Show more

Accomplishments/Portfolio

天体写真

Awards and Certifications

RecSys Challenge 2020: 3rd (prize winning)

June 2020

Kaggle Google QUEST Q&A Labeling: 23th / 1571 (silver)

Feb 2020

Kaggle Days Tokyo Competition: 5th / 88

Dec 2019

atma杯 オンサイトデータコンペ #2 5th / 39

Nov 2019

Kaggle IEEE-CIS Fraud Detection Competition: 5th / 6831(gold)

Oct 2019

Show more


Languages

Japanese - Native