| 中川 廣大 | 1451203: M, 2回目発表 | ソフトウェア設計学 | 飯田 元,松本 健一,安本 慶一,市川 昊平,崔 恩瀞 |
|
title:An Observation of Agile Software Development Process for its improvement
abstract:One of the problems in Agile Software Development is that it is difficult to improve the software process quantitatively. In this research, I examine how to identify the part that developers should improve in Agile Software Development in order to solve the problem. In examining the method, I analyzed Cloud Spiral's exercise data and found two characteristic points in the software process. I propose to perform microprocess analysis, which model process, extract, measure, and evaluate process instance in these two characteristic points. language of the presentation: Japanese 発表題目:アジャイル型ソフトウェア開発におけるプロセス改善部分の特定方法の検討 発表概要:アジャイル型ソフトウェア開発における課題の一つとして，ソフトウェアプロセスの定量的な改善が難しいことが挙げられる．本研究では，その解決のため，アジャイル型開発の開発プロセスにおいて改善すべき部分の特定方法の検討を行う．手法の検討にあたり，Cloud Spiralの演習データを分析し，ソフトウェアプロセスにおける特徴的な点を2つ発見した．この2つの特徴的な点において，マイクロプロセス分析を行い，プロセスをモデル化し，プロセスインスタンスの抽出，計測，評価を行うことを提案する． | |||
| 吉岡 大輝 | 1651120: M, 1回目発表 | ロボティクス | 小笠原 司 |
|
title:Development of meal support system with robot arm attached to tremor suppressing function
abstract:Essential tremor and Parkinson's disease patients become difficult to activities of daily life such as meal action. Therefore, it is necessary to relieve tremor by external devices and exoskeleton to support movement. In this research, we aim to develop a meal support system that attaches the tremor suppressor to the tip of the robot arm, and attaches it to the forearm to reduce the tremor of the hand. In this report, we explain the research background and the problems of the previous research and the proposed method. language of the presentation:Japanese 発表題目:振戦抑制機能を持つロボットアームによる食事支援システムの開発 発表概要:本態性振戦やパーキンソン病といった振戦患者は食事動作などの様々な日常動作が困難なものとなる．よって，振戦を外部装置や外骨格によって軽減して動作を補助することが必要とされている．本研究ではロボットアームの先端に振戦抑制装置を取り付け，それを前腕に装着することによって安定して食事を行うことが出来るシステムの開発を目的とする．本報告では研究背景と先行研究の問題点，提案手法について説明する． | |||
| 鈴木 啓大 | 1651063: M, 1回目発表 | 数理情報学 | 池田和司☆ |
|
title: Study of optimization method for tractography algorithm using functional information
abstract: The cerebral cortex is composed of gray matter and whiter matter. A tool to estimate trajectories of brain white matter using T1 weighted image and diffusion weighted image(DWI) from MRI is called tractography. The accuracy of it was suspected by some results of tracer studies. Although the optimization method improved tractography using DWI information, it suggested the limits of improvement using additional structural information. To overcome it, this research will improve tractography algorithm using functional information from MEG and fMRI. language of the presentation: Japanese 発表題目: 機能的情報を用いたtractographyアルゴリズムの精度向上に関する検討 発表概要: 大脳皮質は神経細胞から成る灰白質と異なる脳領域同士を繋ぐ白質から構成されており，MRIから得られたT1強調画像と拡散強調画像を用いて推定された白質線維束の走行をtractographyという．tractographyは実際の白質線維を調べるtracer研究との比較により精度が問題視されている．先行研究では拡散強調画像から得られた情報を最適化する手法が提案されているが，構造的情報の追加による精度向上には限界があることが示唆されている．そこで本研究ではMEGやfMRIから得られた機能的情報を用いてtractographyアルゴリズムの改善を目指す． | |||
| 米川 柾 | 1651126: M, 1回目発表 | 数理情報学 | 池田和司☆ |
|
title: Age-related difference in functional differentiation of cortico-striatal circuits with behavioral learning using multimodal MRI
abstract: Cognitive function is the ablitily about intellectual activity which includes attention, memory, and planning, and many studies have been reported that cognitive function decreases with the aging. As a brain region related to cognitive function, there is a striatum which is one of the component of the basal ganglia. In previous study, it is known that the striatum is functionally differentiated for different kinds of behavior and the different brain region depends on the stage of action learning. In this study, the relationship between the change in functional differentiation of striatum and age will be clarified by MRI analysis. language of Japanese 発表題目: マルチモーダルMRIを用いた行動学習に伴う大脳皮質-線条体ループの機能分化と年齢との関係 発表概要: 注意力や記憶力、実行機能などの人の知的活動に関わる能力のことを認知機能といい、認知機能は高齢化に伴って低下していくことがこれまで多く報告されてきた。 認知機能に関わる脳部位として、大脳基底核の構成要素の一つである線条体がある。 これまでの研究から、線条体は異なる種類の行動に対して機能分化しており、行動学習の段階によってどの脳領域で処理されるかが変化するという事が知られている。 本研究では、線条体の機能分化の変化と年齢との関係を、MRI解析を用いて明らかにする。 | |||
| 峰友 佑樹 | 1651103: M, 1回目発表 | 光メディアインタフェース | 向川 康博 |
|
title:Measurement of Scattering Phase Function Based on Approximation Model of Single Scattering
abstract: The scattering phase function which represents distribution of scatteredlight is an important factor in rendering photo-realistic computer graphics. Conventional methods require special setup or many observations for the measurement. In this research, we obtain the scattering phase function by simple setup based on approximation model of single scattering. First, to eliminate the influence of multiple scattering light on observation, we constructed experimental setup using polarizer. Next, in order to validate the effect of multiple scattering light removal, we measure the scattering light from several scattering medium of different concentration. We analyze and evaluate this measurement result. language of the presentation: Japanese | |||
| 萬代 光治 | 1651101: M, 1回目発表 | 大規模システム管理 | 笠原 正治 |
|
title: Analysis of cloud platform workload data
abstract: Demand for cloud platforms such as Google Cloud Platform and Azure is increasing, so improvement of scheduling efficiency of jobs is an important task. However, on large-scale platforms, the mixture of various jobs and hardware makes job/task scheduling complicated. In order to promote research on the nature and challenges of scheduling, Google released Google Cluster Usage Trace, which is the trace data of a platform actually operated. In this presentation, I will introduce the paper which analyzes Google Cluster Usage Trace and presents properties such as the heterogeneity of machines constituting the platform and the dynamics of job submission. Next, I will show the results of my analysis of how jobs submitted to the platform were scheduled. language of the presentation: Japanese 発表題目: クラウドプラットフォームのワークロードデータの解析 発表概要: Google Cloud PlatformやAzureに代表されるクラウドプラットフォームの需要は高まりつつあるため、ジョブのスケジューリング効率の改善は急務である。しかしながら大規模なクラウドプラットフォームでは、多様なジョブやハードウェアの混在などがスケジューリングを複雑にすることがある。スケジューリングの性質や課題についての研究を促進するため、Google社は実際に運用されたプラットフォームのトレースデータであるGoogle Cluster Usage Traceを公開した。本発表ではまず、Google Cluster Usage Traceを解析してプラットフォームを構成する機器の不均質さやジョブ投入の動態などの性質を提示した論文の紹介を行う。次にプラットフォームに投入されたジョブがどのようにスケジューリングをされたかについて独自で解析した結果について述べる。 | |||
| 増田 健一 | 1551101: M, 2回目発表 | モバイルコンピューティング | 伊藤 実,藤川 和利,柴田 直樹,川上 朋也 |
|
title: Study of a Presupposed Road Pricing System to Reduce Traffic Congestion
abstract: Traffic congestion causes many problems, for example, an economic loss and environmental pollution. To reduce an influence of the traffic congestion, road pricing schemes have been proposed and implemented. Variable tolls (road prices) based on traffic conditions are effective for high efficiency of road utilization, however, the users in the existing road pricing methods cannot know the prices to pay actually because of the difficulty of predicting traffic conditions. This research aims to provide comfortable and stable traffic environments without traffic congestion, and we propose a method to decide a toll in advance of using roads by the departure point, arrival point, and using time. In this presentation, I introduce the existing road pricing methods and those problems. After that, I explain my proposing method for a presupposed road pricing scheme. language of the presentation: Japanese | |||
| 中村 優吾 | 1551068: M, 2回目発表 | ユビキタスコンピューティングシステム | 安本 慶一,藤川 和利,荒川 豊,諏訪 博彦,藤本 まなと |
|
title : Study of Local IoT Platform for Proximity Distributed Real-time Processing of IoT Data Flows abstract : Edge Computing and Fog Computing are new paradigms where data processing is executed in or on the edge of networks to mitigate cloud server load. However, Edge Computing and Fog Computing still need powerful servers on the edge of networks which impose additional costs for deployments. We proposed a platform called IFoT (Information Flow of Things) that efficiently performs distributed processing as well as distribution and analysis of data streams near their sources based on "Process On Our Own (PO3)'' concept. In IFoT, processing of tasks for cloud servers is delegated to an ad-hoc distributed system consisting of proximity IoT devices for distributed real-time stream processing. In this presentation, aiming to realize IFoT concept, we show the design of IFoT middleware for IoT devices where those devices process data streams in real-time and in a distributed manner and the results of the feasibility study.
| |||
| 新垣 杏里 | 1551004: M, 2回目発表 | 情報基盤システム学 | 藤川 和利,安本 慶一,笠原 正治,新井 イスマイル,猪俣 敦夫 |
|
title: *** replace this part with the title of your talk IN ENGLISH ***
abstract: *** replace this part with the abstract of your talk IN ENGLISH *** language of the presentation: *** English or Japanese (choose one) *** | |||
| 吉川 将司 | 1551119: M, 2回目発表 | 自然言語処理学 | 松本 裕治,中村 哲,新保 仁,進藤 裕之,能地 宏 |
|
title: Joint Transition-based Dependency Parsing and Disfluency Detection for Automatic Speech Recognition Texts
abstract: Joint dependency parsing with disfluency detection is an important task in speech language processing. Recent methods show hight performance for this task, although most authors make the unrealistic assumption that input texts are transcribed by human annotators. In real-world applications, the input text is typically the output of an automatic speech recognition (ASR) system, which implies that the text contains not only disfluency noises but also recognition errors from the ASR system. In this work, we propose a parsing method that handles both disfluency and ASR errors using an incremental shift-reduce algorithm with several novel features suited to ASR output texts. Because the gold dependency information is usually annotated only on transcribed texts, we also introduce an alignment-based method for transferring the gold dependency annotation to the ASR output texts to construct training data for our parser. We conducted an experiment on the Switchboard corpus and show that our method outperforms conventional methods in terms of dependency parsing and disfluency detection. language of the presentation: Japanese | |||
| 富士 秀 | 1561018: D, 中間発表 | 自然言語処理学 | 松本 裕治,中村 哲,新保 仁,進藤 裕之,能地 宏 |
| 澤山 熱気 | 1651056: M, 1回目発表 | 自然言語処理学 | 松本 裕治 |
|
title: Extract the knowledge from the paper
abstract: Looking over a large number of papers when gathering related papers is costly to both quantitative and temporal. Especially, it is difficult to find out papers from non-maintained domain papers. In this research, focusing on material domains, we extract substances, attributes and their relationships to construct a knowledge extraction system using small annotated data. In this presentation, I will introduce the background, the proposed method and current progress. language of the presentation: Japanese 発表題目: 論文からの知識抽出 発表概要: 関連する論文を収集する際に大量の論文に目を通すことは，量的，時間的に大きなコストがかかる．特に，整備されていないドメインにおいて論文を探し出すことが困難となっている．そこで本研究では物質ドメインに着目し，知識抽出システムの構築のために，人手でアノテーションされた少量のデータを用いて，物質，属性とそれらの関係を抽出することを目的とする．本報告では研究背景，提案手法ならびに現在の研究状況について報告する． | |||
| 河野 誠也 | 1651038: M, 1回目発表 | 知能コミュニケーション | 中村 哲 |
|
title: Dialogue System for Information Retrieval based on Dialogue Acts in Reference Interview
abstract: Questions expressing user’s information demands are often ambiguous in a dialogue system for information retrieval. The ambiguous queries make it difficult for the system to provide the information that the user really needs. It is important to clarify the users' question by using some actions from the system such as asking again or recommending query candidates. The reference service is known as a service to seek information at library on a human-to-human basis. In this study, we focus on “reference interviews” that are conversations between librarians and users to clarify users’ true request in the reference service. We analyze the process for information (document) retrieval between librarians and library users and model the response strategy to clarify the user's true request in dialog system. language of the presentation: Japanese 発表題目: レファレンス・インタビューにおける対話行為を用いた情報検索対話システム 発表概要: 情報検索をタスクとする対話システムにおいて，ユーザが自身の情報要求を表現したクエリの意図が曖昧となり、そのクエリのみでは応答することが難しいことがある．この場合，システム側から何らかの聞き返しや情報提示を行いユーザの質問意図の明確化を行うことが重要となる．人対人の対話で情報検索をタスクとするものとして，図書館におけるレファレンス・サービスがある．本研究では，レファレンス・サービスにおいて，図書館員が図書館利用者の曖昧な要求を明確化することを目的とし行うレファレンス・インタビューに着目する．具体的には、図書館員と図書館利用者との間で行われる情報探索プロセスについて分析を行い，対話システムにおけるユーザの質問意図明確化のための応答戦略のモデル化への応用について検討を行う． | |||