Hiroki Tanioka

Hiroki Tanioka
Tokushima University, Tokushima, Japan · Center for Administration of Information Technology

Ph.D

About

35
Publications
3,001
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
44
Citations
Introduction
Additional affiliations
April 2016 - present
Tokushima University
Position
  • Professor (Assistant)
September 2014 - March 2016
Works Applications
Works Applications
Position
  • Researcher
April 2011 - September 2014
FITEC Corporation
Position
  • R&D Manager

Publications

Publications (35)
Conference Paper
This paper proposes a novel approach for retrieving large-scale XML data using the vector space model. The vector space model is commonly used in the information retrieval community. Last year, for the Evaluation of XML Retrieval (INEX) 2006 Adhoc Track, we developed a system using fragment elements. The system made it possible to search over XML...
Conference Paper
In this paper, we propose an improved information retrieval model, where the integration of modification-words and head-words is introduced into the representation of user queries and the traditional vector space model. We show how to calculate the weights of combined terms in vectors. We also propose a new strategy to construct the thesaurus in a...
Conference Paper
Full-text available
We developed a passage retrieval system for XML documents using the vector space model. To be more flexible for the query, we also developed a method of unification of multiple retrieved elements and a fragment indexing system. Our system is composed of an inverted file and an XML Path Language (XPath) path list. The validity of the method was test...
Article
Full-text available
We developed a patent retrieval system with the cor-responding very large number of patents from NTCIR-6 Patent Retrieval Task. And we developed a method of refining and emphasizing query. Our retrieval system consisting of four PCs could make indices of all claims in specifications for ten years. Then we confirmed that the query emphasis was bette...
Conference Paper
This paper reports the result of experimentation of our approach using the vector space model for retrieving large-scale XML data. The purposes of the experiments are to improve retrieval precision on the INitiative for the Evaluation of XML Retrieval (INEX) 2008 Adhoc Track, and to compare the retrieval time of our system to other systems on the...
Article
Full-text available
In Japan, programming education has been made compulsory in elementary schools since 2020. The Programming Education Guide (GPE) explains the purpose of programming education and the abilities that can be fostered through programming education. In addition, the “Portal Site for Programming Education Focusing on Elementary Schools” introduces variou...
Article
Full-text available
Tokyo 2020 Olympic Games has been postponed until 2021. Most of the 33 sports still planned for the Olympic Games in 2021 will use data. The sports data gathered using various method is analyzed by experts. The experts also called sports data analysts have been developed various systems and methods using the sports data and Artificial Intelligence...
Preprint
Full-text available
Fast and scalable Content-Based Image Retrieval using visual features is required for document analysis, Medical image analysis, etc. in the present age. Convolutional Neural Network (CNN) activations as features achieved their outstanding performance in this area. Deep Convolutional representations using the softmax function in the output layer ar...
Poster
Full-text available
This poster presents a progress of our development for a player tracking system using a 360 degree camera. There are already some tracking systems using GPS, RFID, and Wireless LAN. There are also a multi-camera tracking system and even a tracking system with a single-camera.
Poster
Full-text available
There are large-scale English question-answer corpora available. However, there are no large-scale Japanese question-answer corpus. Furthermore, various types of question-answer corpora are needed for each domain. Hence, an automatic generation method of question-answer pairs is valuable. Our approach is composed of three steps. The first step is t...
Conference Paper
Full-text available
Luck or unluck is supposed to exist in baseball. In this paper, we consider unpredictable "lucky" or "un-lucky" cases which are occurred on the ground in order to find out authentic ability of batters. If all cases are observable, hit probabilities can be calculated. However, it is impossible to gather all the information on the ground. Therefore,...
Presentation
Full-text available
This paper provides detailed suggestions to create an Image Search Engine with Deep Learning. There are still few attempts with Deep Learning on a search engine. Here is a good idea of an extremely easy way of building an image search with Elasticsearch and Keras on Jupyter Notebook. So, it is demonstrated how an image search engine can be created...
Poster
Full-text available
This paper provides a case study of improving availability for searching documents in an academic facility. Every day some documents are needed for our business and academic activities in the Center for Administration of Information Tech- nology (AIT) in Tokushima University. AIT has been employing and operating Information Security Management Syst...
Conference Paper
Currently, authentication methods using ID and password are widely used and fulfilled central roles in various information systems and services. Our university also uses ID and password for authentication of most services. However, passwords have various problems such as reuse, phishing and leakage. This research is a practical experiment in order...
Article
Full-text available
徳島大学情報センターは, ISMS に基づく情報セキュリティポリシーに則り, 教員及び職員が作成 した ISMS 文書をファイルサーバで管理している. ISMS 文書以外の本センターが関わる業務文書, 契約書, マニュアル, ログ等といった業務運用系文書についても, 同一のファイルサーバで管理して いる状況である. ISMS 文書については, ディレクトリ構造やファイル名に運用規定を設けることに よって, 必要な人が必要なときに使用できる状態を維持している. しかしながら, 教員及び職員全員 が, ファイルサーバのディレクトリの最新状況を常に把握することは困難なため, ISMS 文書やその 他の必要書類を即座に使用できない場面があるのも事実である. この状況を改善するため, 我々は, 本セン...
Article
Full-text available
We developed a distributed search system with the corresponding very large scale corpora from NTCIR-5 WEB Task. And we arranged the scoring method which is based on link-structure of the Web documents to calculate lower cost. Our search system, which con-sists of 6 PCs could make indices for full texts size of about 1 TB. Additionally, we confirmed...
Article
Full-text available
本稿では,従来法の1 つであるベイジアンフィルタを用いたspam メールフィルタの精度(true negative rate)を改善する方法について提案する.これまでの学習型spam メールフィルタとしては,ベイジアンフィルタがよく利用されており,一定の成果が得られている.しかしながら,ベイジアンフィルタを利用した方法においても,誤検出率(false positive rate)の低減や,さらなる精度向上が期待される.我々は,単語のspam 確率(尤度)の分布およびメールのspam 度の分布状況を分析し,誤検出をおさえながらも,高い判定精度を実現する方法について提案し,その精度について,従来方式と比較して評価する.We propose an improved baysian filter f...
Conference Paper
近年,さまざまな場面で利用されようになった XML 文書に対する検索の要求は,日増しに増加している.XML 文書に対する検索のアプローチには,データベース分野によるものと,情報検索分野によるものの大きく分けて 2 つのアプローチがあるが,我々は,情報検索分野で広く知られているベクトル空間法を利用し,断片化された検索結果をユーザニーズに合わせて効率的にマージする手法について検討している.この方法では,転置リストの再構築なしにさまざまなスコアリングを試すことができるが,XML ノードである検索結果をマージする際に,ノードの親子関係を判定しながらマージを行う処理時間に改善の余地があった.本稿では,相対逆経路リストを用いることにより,これを改善する.さらに,システムの有効性について,The Init...
Technical Report
学習型spamフィルタのより精密な評価を行うため,新たな精度指標である総合エラー率を提案する.この指標により,既存の精度指標である追加学習性能・学習収束性能・再現率などの静的・動的な精度指標が統合でき,精度比較やチューニングの自動化が可能となる.また,この精度指標に基づいて,実際にspamフィルタのパラメータをチューニングし,指標の有用性を実証した. Total error rate is a novel accuracy measure to evaluate spam filters more precisely. This measure is the integration of existing static or dynamic accuracy measures such a...
Technical Report
近年, 増加傾向にあるspamメールへの対策として, メールソフトによるspamフィルタがよく用いられている.しかし, spamメールの送信者や内容は日々変化するため, より高精度な学習型spamフィルタが求められている.本稿では, 学習型spamフィルタを実現するため, 静的な判定精度に加えて, 動的な判定精度を評価する手法について紹介する.この手法は, 評価項目を初期性能, 追加学習性能, 学習収束性能に分けて評価することで, ユーザビリティの観点からも適切な学習型spamフィルタを実現し, ビジネス上の実用性を高めることを目的とする. We propose evaluation methods for improvement of sensory accuracy in spam m...
Technical Report
近年, インターネット技術を基盤とした電子メールやWWWの普及に伴って, 文章を構成する文字列パターンには, 言語的な意味を持つ単語以外に, 顔文字や絵文字といったアスキーアートを用いたものが多く見られるようになってきた.アスキーアートは通常, 言語的な意味を持たず, 文字の配置と組み合わせによって, 視覚的な情報として読み手に意味を伝える.このため, 従来の形態素解析に代表される自然言語処理技術では, 正確な意味理解ができない.本研究では, 文章からアスキーアートを正しく抽出するために, Support Vector Machineを用いて高精度にアスキーアートを識別する方法を提案する. Late years, ASCII Art is commonly found among Inte...

Network

Cited By

Projects

Projects (7)
Project
Programming education about computational thinking and computer science.
Project