タスク概要・参加者募集

第13回 NTCIR (2016 - 2017)
情報アクセス技術研究のためのテストベッドとコミュニティ
カンファレンス: 2017年12月 5-8日 東京 学術総合センター


NTCIR-13 タスク参加のご案内:

Call for Participation [Flyer]

参加申込の手引き

情報アクセス技術向上のための協同的な取り組みに参加してみませんか?

第13回目のNTCIR、NTCIR-13では、共通のデータセットを用いて研究するタスクへの参加チームを募集中です。
情報アクセス技術の評価には、研究者の協同作業の結果として作成される「テストコレクション」に基づく評価が欠かせません。NTCIRは、数多くの研究者の協力の下で、その評価基盤の形成に過去15年以上に渡って取り組み、技術の発展に貢献してきました。そして日々開発される新しい技術に対する評価手法を模索しつつ、活動を進めております。
情報アクセス分野の学生や若手研究者のみなさん,先生方,企業で研究をなさっている方,または情報学に興味のある方々,大規模なテストコレクションを用いた検索、質問応答、自然言語処理に
関心のある研究グループは、どなたでも歓迎します。 どうぞ、奮ってご参加ください。

参加登録はこちらをご覧ください:http://research.nii.ac.jp/ntcir/ntcir-13/howto-ja.html

NTCIRについて

評価タスク

第13回NTCIR(NTCIR-13)タスク選考委員会は、以下の5つのコアタスクと4つのパイロットタスクを選定しました。
タスク紹介スライド(キックオフイベント)を下記のページからご覧いただけます:
タスク紹介スライド(キックオフイベント): http://research.nii.ac.jp/ntcir/ntcir-13/kickoff-ja.html
タスクの詳細・最新情報について、下記のタスク概要および各タスクのウェブサイトをご覧ください。

Lifelog-2    MedWeb    OpenLiveQ    QALab-3    STC-2    AKG    ECA    NAILS    WWW    

コアタスク

Personal Lifelog Organisation & Retrieval Task ("Lifelog-2")

"多様なデバイスを用いて蓄積したマルチモダルライフログデータの検索と要約 "

Abstract:
Personal lifelogging is the process of capturing multiple aspects of one's life in digital form, and is set to become an increasingly important aspect of people's lives. The main objectives of this Lifelog-2 task are to build upon the momentum of the NTCIR-12 lifelog task, to develop a new (more semantically rich) test collection, and to encourage collaborative research in this field within the wider multimedia analytics and IR communities by organising useful and novel research challenges. To this end, we are proposing the continuation of two sub-tasks from NTCIR-12 (LSAT and LIT) and include one new task (LET). The LSAT task will continue to advance community capabilities for accessing lifelogs by exploring search challenges, the LIT task will take a novel and attractive view of lifelog insights generation, while the LET task focuses on exploring approaches to semantic enrichment of raw lifelog data.

Website: http://ntcir-lifelog.computing.dcu.ie/
Contact:

PAGE TOP


Medical Natural Language Processing for Web Document ("MedWeb")

"Twitterの発言テキストと闘病記ブログに対する病名ラベリングおよび病名抽出"

Abstract:
Recently, increasing number of medical records has been written as the form of electronic media replacing paper media, thereby digital information processing in medical fields has been increasingly needed. Nowadays, this trend focuses not only on electronic health records but also on various patients’ texts, such as social media texts, web blogs, and so on. In the NTCIR-13 MedWeb Task, participants are required to extract disease information from two types of patient texts in Japanese, English, and Chinese; (1) Twitter message texts and (2) disease journal texts. Since this task setting can be formalized as labeling or extracting disease names of these medical texts, the achievements of this task can be almost directly applied to a fundamental engine for actual applications.

Website: http://mednlp.jp/medweb/NTCIR-13/
Contact:

PAGE TOP


Open Live Test for Question Retrieval ("OpenLiveQ")

"Yahoo!知恵袋の実サービス環境にて評価が行われる質問検索タスク"

Abstract:
Open Live Test for Question Retrieval (OpenLiveQ)タスクでは,質問検索システムの評価のために,ヤフー株式会社のコミュニティQ&Aサービスにおけるオープンライブテスト環境を提供します.このタスクでは,より現実的なシステム評価を行う機会を提供し,実検索サービスに特有の問題(e.g. クエリの曖昧性・不明瞭性,多様な適合性基準)に取り組むことを支援します.「クエリと回答付き質問集合が与えられたときに順位付きの質問リストを返す」タスクとなっています.

Website: http://www.openliveq.net/
Contact:

PAGE TOP


QA Lab for Entrance Exam ("QALab-3")

"実世界質問応答。対象は論述問題を含む大学入試(世界史)"

The goal is to investigate the real-world complex Question Answering (QA) technologies using Japanese university entrance exams and their English translation on the subject of "World History". The questions were selected from two different stages - The National Center Test for University Admissions (multiple choice-type questions) and from secondary exams at multiple universities (complex questions including essays). All the questions are provided in an XML format.

Website: http://research.nii.ac.jp/qalab/
Contact: qalab-admin

PAGE TOP


Short Text Conversation("STC-2")

"ユーザの発言に対し、システムは適切な応答を検索できるか?生成できるか?"

Abstract:
At NTCIR-12, STC is taken as an IR problem by maintaining a large repository of post-comment pairs from Weibo in Chinese subtask and Twitter in Japanese subtask, and then finding a clever way to reuse these existing comments to respond to new posts. At NTCIR-13, besides the retrieval- based method, we propose to add a new subtask called generation-based method to generating “new” comments. The generation-based method has emerged as a hot research topic and gained the most attention in recent years, while it is still an open problem whether the IR-based method should be wholly replaced by or combined with generation-based method for STC task. By organizing this task at NTCIR-13, we will provide a transparent platform to compare the two aforementioned methods via doing comprehensive evaluations. Furthermore, participants are encouraged to explore some effective ways to combine the two methods to get a more intelligent chatbot in the future runs of STC.

Website:
http://ntcirstc.noahlab.com.hk/STC2/stc-cn.htm
Contact:

PAGE TOP


パイロットタスク

Actionable Knowledge Graph ("AKG")

"行動的な検索意図のためのナレッジグラフ生成"

Abstract:
Knowledge graph has become increasing common and important component in search engine result pages (SERPs). However, the current knowledge graph remains informational and do not work with complex queries. This task aims to facilitate the development of new technologies to create “actionable” (or transactional) knowledge graph presentation that can be used for query intents such as buying event tickets, listening to music online or reserving a restaurant. In this pilot task, we set two subtasks: Action mining task and actionable knowledge graph generation task.

Website: http://ntcirakg.github.io/index.html
Contact:

PAGE TOP


Emotion Cause Analysis ("ECA")

"ニュース文書から感情およびその原因を抽出"

Abstract:
In recent years, there has been a large body of research work on emotion classification from text. However, in many cases, we care more about the stimuli, or the cause of emotions. Unfortunately, the lack of annotated corpora and standard metrics for this task has limited the research on this topic. Thus, we propose to organize a new task, emotion cause analysis (ECA), in NTCIR 13 to evaluate the state-of-the-art emotion cause extraction/detection methods on Chinese and English news text. We set two subtasks: 1.detecting the clause which contains emotion cause, and 2. Detecting the exact boundary of the emotion cause.

Website: http://hlt.hitsz.edu.cn/ECA.html
Contact:

PAGE TOP


Neurally Augmented Image Labelling Strategies ("NAILS")

"脳波を用いた画像内容のラベル付け"

Abstract:
EEG (Electroencephalography) has recently become an accessible method for researchers and users to build and operate BCI (Brain-Computer Interface) applications, primarily because of the availability of a new generation of low-cost devices. While the initial use of such techniques began in clinical/rehabilitative settings for the purposes of augmenting communication and control, a recent trend has been to use such signals and methods in new domains, such as the image annotation task, which relies on the identification of target brain events to trigger labeling.
This trend is particularly relevant to the IR community as in recent years EEG has become a promising technology for the purposes of annotating multimedia content, or identifying when a user’s attention is drawn to something in the real world, or even as a source of user sensor data to be indexed for later retrieval. However, working with EEG data is challenging and contains many pitfalls for inexperienced researchers.
Our task will explore better ways to annotate a collection of multimedia data, introduce IR researchers to EEG data, allow them to explore the application of EEG data in a controlled experimental way, and produce the first EEG in IR test collection.
Website: http://ntcir-nails.computing.dcu.ie/
Contact:

PAGE TOP


We Want Web ("WWW")

"還ってきたウェブ検索タスク - 少なくとも3回のNTCIRに渡る評価を行い、技術進歩を定量化します"

Abstract:
We would like to run an ad hoc Web task for at least three rounds at NTCIR, and evaluate systems not only with traditional measures but also with more advanced measures that better reflect user experiences. For some of our test topics, we will provide user behaviour data to participants so that they can tune their systems for the latter type of measures. Participants will try to go beyond the IR performance plateau; We organisers will establish statistically-motivated methods for designing and maintaining test collections and for quantifying the progress across NTCIR rounds. Both parties will collectively conduct failure analysis for the next NTCIR rounds. WWW would use the up-to-date Sogou-Q-16(querylog) and Sogou-T-16(corpus) to promote researches.
Website: http://www.thuir.cn/ntcirwww
Contact:

PAGE TOP



Last modified: 2016-10-14