scholarly journals The NarrativeQA Reading Comprehension Challenge

2018 ◽  
Vol 6 ◽  
pp. 317-328 ◽  
Author(s):  
Tomáš Kočiský ◽  
Jonathan Schwarz ◽  
Phil Blunsom ◽  
Chris Dyer ◽  
Karl Moritz Hermann ◽  
...  
Keyword(s):  
Information Retrieval ◽  
Reading Comprehension ◽  
Pattern Matching ◽  
Question Answering ◽  
Local Context ◽  
Artificial Agents ◽  
Learning To Read ◽  
Term Frequency ◽  
Context Similarity

Reading comprehension (RC)—in contrast to information retrieval—requires integrating information and reasoning about events, entities, and their relations across a full document. Question answering is conventionally used to assess RC ability, in both artificial agents and children learning to read. However, existing RC datasets and tasks are dominated by questions that can be solved by selecting answers using superficial information (e.g., local context similarity or global term frequency); they thus fail to test for the essential integrative aspect of RC. To encourage progress on deeper comprehension of language, we present a new dataset and set of tasks in which the reader must answer questions about stories by reading entire books or movie scripts. These tasks are designed so that successfully answering their questions requires understanding the underlying narrative rather than relying on shallow pattern matching or salience. We show that although humans solve the tasks easily, standard RC models struggle on the tasks presented here. We provide an analysis of the dataset and the challenges it presents.

scholarly journals Translucent Answer Predictions in Multi-Hop Reading Comprehension

2020 ◽  
Vol 34 (05) ◽  
pp. 7700-7707
Author(s):  
G P Shrivatsa Bhargav ◽  
Michael Glass ◽  
Dinesh Garg ◽  
Shirish Shevade ◽  
Saswati Dana ◽  
...  
Keyword(s):  
Reading Comprehension ◽  
Question Answering ◽  
State Of The Art ◽  
Local Context ◽  
The Novel ◽  
Loose Coupling ◽  
Loosely Coupled ◽  
Neural Architecture ◽  
Coupled Networks ◽  
Novel Design

Research on the task of Reading Comprehension style Question Answering (RCQA) has gained momentum in recent years due to the emergence of human annotated datasets and associated leaderboards, for example CoQA, HotpotQA, SQuAD, TriviaQA, etc. While state-of-the-art has advanced considerably, there is still ample opportunity to advance it further on some important variants of the RCQA task. In this paper, we propose a novel deep neural architecture, called TAP (Translucent Answer Prediction), to identify answers and evidence (in the form of supporting facts) in an RCQA task requiring multi-hop reasoning. TAP comprises two loosely coupled networks – Local and Global Interaction eXtractor (LoGIX) and Answer Predictor (AP). LoGIX predicts supporting facts, whereas AP consumes these predicted supporting facts to predict the answer span. The novel design of LoGIX is inspired by two key design desiderata – local context and global interaction– that we identified by analyzing examples of multi-hop RCQA task. The loose coupling between LoGIX and the AP reveals the set of sentences used by the AP in predicting an answer. Therefore, answer predictions of TAP can be interpreted in a translucent manner. TAP offers state-of-the-art performance on the HotpotQA (Yang et al. 2018) dataset – an apt dataset for multi-hop RCQA task – as it occupies Rank-1 on its leaderboard (https://hotpotqa.github.io/) at the time of submission.

UTILIZING LOCAL CONTEXT FOR EFFECTIVE INFORMATION RETRIEVAL

2008 ◽  
Vol 07 (01) ◽  
pp. 5-21 ◽  
Author(s):  
TANVEER J. SIDDIQUI ◽  
UMA SHANKER TIWARY
Keyword(s):  
Information Retrieval ◽  
Similarity Measure ◽  
Question Answering ◽  
Graph Matching ◽  
Information Filtering ◽  
Similarity Measures ◽  
Local Context ◽  
Structural Variations ◽  
Retrieval Effectiveness ◽  
Relational Similarity

Our research focuses on the use of local context through relation matching to improve retrieval effectiveness. An information retrieval (IR) model that integrates relation and keyword matching has been used in this work. The model takes advantage of any existing relational similarity between documents and query to improve retrieval effectiveness. It gives high rank to a document in which the query concepts are involved in similar relationships as in the query, as compared to those in which they are related differently. A conceptual graph (CG) representation has been used to capture relationship between concepts. A simplified form of graph matching has been used to keep our model computationally tractable. Structural variations have been captured during matching through simple heuristics. Four different CG similarity measures have been proposed and used to evaluate performance of our model. We observed a maximum improvement of 7.37% in precision with the second CG similarity measure. The document collection used in this study is CACM-3204. CG similarity measure proposed by us is simple, flexible and scalable and can find application in many IR related tasks like information filtering, information extraction, question answering, document summarization, etc.

Rancang Bangun Aplikasi Chatbot Sebagai Media Pencarian Informasi Anime Menggunakan Regular Expression Pattern Matching

2017 ◽  
Vol 9 (1) ◽  
pp. 19-24 ◽  
Author(s):  
David Domarco ◽  
Ni Made Satvika Iswari
Keyword(s):  
Information Retrieval ◽  
Expression Pattern ◽  
Pattern Matching ◽  
Language Processing ◽  
Regular Expression ◽  
Technology Development ◽  
Data Retrieval ◽  
Index Terms ◽  
Retrieval Engine ◽  
Behavioral Intention To Use

Technology development has affected many areas of life, especially the entertainment field. One of the fastest growing entertainment industry is anime. Anime has evolved as a trend and a hobby, especially for the population in the regions of Asia. The number of anime fans grow every year and trying to dig up as much information about their favorite anime. Therefore, a chatbot application was developed in this study as anime information retrieval media using regular expression pattern matching method. This application is intended to facilitate the anime fans in searching for information about the anime they like. By using this application, user can gain a convenience and interactive anime data retrieval that can’t be found when searching for information via search engines. Chatbot application has successfully met the standards of information retrieval engine with a very good results, the value of 72% precision and 100% recall showing the harmonic mean of 83.7%. As the application of hedonic, chatbot already influencing Behavioral Intention to Use by 83% and Immersion by 82%. Index Terms—anime, chatbot, information retrieval, Natural Language Processing (NLP), Regular Expression Pattern Matching

scholarly journals Multi-step Entity-centric Information Retrieval for Multi-Hop Question Answering

2019 ◽  
Author(s):  
Rajarshi Das ◽  
Ameya Godbole ◽  
Dilip Kavarthapu ◽  
Zhiyu Gong ◽  
Abhishek Singhal ◽  
...  
Keyword(s):  
Information Retrieval ◽  
Question Answering ◽  
Entity Centric

scholarly journals Evaluating Commonsense in Pre-Trained Language Models

2020 ◽  
Vol 34 (05) ◽  
pp. 9733-9740 ◽  
Author(s):  
Xuhui Zhou ◽  
Yue Zhang ◽  
Leyang Cui ◽  
Dandan Huang
Keyword(s):  
Reading Comprehension ◽  
Question Answering ◽  
Deep Level ◽  
Language Models ◽  
Future Research ◽  
Correct Prediction ◽  
Test Cases ◽  
Word Sense ◽  
Training Set ◽  
Text Data

Contextualized representations trained over large raw text data have given remarkable improvements for NLP tasks including question answering and reading comprehension. There have been works showing that syntactic, semantic and word sense knowledge are contained in such representations, which explains why they benefit such tasks. However, relatively little work has been done investigating commonsense knowledge contained in contextualized representations, which is crucial for human question answering and reading comprehension. We study the commonsense ability of GPT, BERT, XLNet, and RoBERTa by testing them on seven challenging benchmarks, finding that language modeling and its variants are effective objectives for promoting models' commonsense ability while bi-directional context and larger training set are bonuses. We additionally find that current models do poorly on tasks require more necessary inference steps. Finally, we test the robustness of models by making dual test cases, which are correlated so that the correct prediction of one sample should lead to correct prediction of the other. Interestingly, the models show confusion on these test cases, which suggests that they learn commonsense at the surface rather than the deep level. We release a test set, named CATs publicly, for future research.

scholarly journals Sistem Rekomendasi Produk Pena Eksklusif Menggunakan Metode Content-Based Filtering dan TF-IDF

2020 ◽  
Vol 5 (3) ◽  
pp. 229
Author(s):  
Mariani Widia Putri ◽  
Achmad Muchayan ◽  
Made Kamisutara
Keyword(s):  
Information Retrieval ◽  
Customer Relationship Management ◽  
Relationship Management ◽  
Customer Relationship ◽  
Brand Awareness ◽  
Product Knowledge ◽  
Inverse Document Frequency ◽  
Term Frequency ◽  
Document Frequency ◽  
Content Based Filtering

Sistem rekomendasi saat ini sedang menjadi tren. Kebiasaan masyarakat yang saat ini lebih mengandalkan transaksi secara online dengan berbagai alasan pribadi. Sistem rekomendasi menawarkan cara yang lebih mudah dan cepat sehingga pengguna tidak perlu meluangkan waktu terlalu banyak untuk menemukan barang yang diinginkan. Persaingan antar pelaku bisnis pun berubah sehingga harus mengubah pendekatan agar bisa menjangkau calon pelanggan. Oleh karena itu dibutuhkan sebuah sistem yang dapat menunjang hal tersebut. Maka dalam penelitian ini, penulis membangun sistem rekomendasi produk menggunakan metode Content-Based Filtering dan Term Frequency Inverse Document Frequency (TF-IDF) dari model Information Retrieval (IR). Untuk memperoleh hasil yang efisien dan sesuai dengan kebutuhan solusi dalam meningkatkan Customer Relationship Management (CRM). Sistem rekomendasi dibangun dan diterapkan sebagai solusi agar dapat meningkatkan brand awareness pelanggan dan meminimalisir terjadinya gagal transaksi di karenakan kurang nya informasi yang dapat disampaikan secara langsung atau offline. Data yang digunakan terdiri dari 258 kode produk produk yang yang masing-masing memiliki delapan kategori dan 33 kata kunci pembentuk sesuai dengan product knowledge perusahaan. Hasil perhitungan TF-IDF menunjukkan nilai bobot 13,854 saat menampilkan rekomendasi produk terbaik pertama, dan memiliki keakuratan sebesar 96,5% dalam memberikan rekomendasi pena.

scholarly journals PENGEMBANGAN BUKU AUDIO UNTUK MENGATASI KESULITAN BELAJAR ANAK DISLEKSIA

2015 ◽  
Vol 3 (2) ◽  
pp. 91
Author(s):  
Faiza Indriastuti
Keyword(s):  
Reading Comprehension ◽  
Learning Difficulties ◽  
Learning Problems ◽  
Instructional Media ◽  
Learning To Read ◽  
Visual Text ◽  
The One ◽  
Text Images

Difficulty learning for learners refers to significant learning problems in learning. One is dyslexics who have difficulty in reading and reading comprehension. Therefore needed the help of technology that can be used as a tool for dyslexic learners in that learning, so as to overcome gaps in their understanding of learning. This article discusses how to develop instructional media as a solution that can be used to overcome for learners difficulties as dyslexic. The one of technologies development that relevant can be used to help students with dyslexia is audiobooks. DTB is one of audiobooks format that can assist learners with learning difficulties as dyslexics become better learners. Because, DTB is can be an effective aids to support the learning of reading and increase in reading comprehension, so as to improve the ability of learners with dyslexia that will ultimately lead to better of value lessons. DTB form in accordance with the needs of dyslexic learners is Tobi DAISY, which is in the form of digital talking books are synchronized between the visual (text, images, tables, charts) and audio. It is possible to make it easier for dyslexic learners in learning to read or understand the reading. The purpose of this article is to give an overview of Tobi DAISY de-velopment that could be used and produced individually for dyslexics to fit the required content. Through Tobi DAISY advantages, it can be concluded that this relevant to be used for dyslexics to help in reading and reading comprehension. AbstrakKesulitan belajar bagi peserta didik mengacu pada masalah belajar yang signifikan dalam pembelajaran. Salah satunya adalah penderita disleksia yang mempunyai kesulitan dalam membaca maupun memahami bacaan. Oleh karenanya diperlukan bantuan teknologi yang dapat digunakan sebagai alat bantu peserta didik disleksia dalam belajar membaca atau memahami bacaan, sehingga dapat mengatasi kesenjangan pemahaman mereka dalam pembelajaran. Artikel ini membahas tentang bagaimana mengembangkan media pembelajaran sebagai solusi yang dapat digunakan untuk mengatasi kesulitan belajar peserta didik disleksia. Salah satu pengembangan teknologi yang relevan dapat digunakan membantu peserta didik disleksia adalah buku audio. DTB merupakan salah satu format buku audio yang membantu peserta didik yang memiliki kesulitan belajar menjadi pebelajar yang lebih baik. Karena, DTB dimungkinkan dapat menjadi alat yang efektif untuk mendukung dalam kegiatan belajar membaca dan peningkatan pemahaman bacaan, sehingga dapat meningkatkan kemampuan peserta didik disleksia dalam membaca dan memahami bacaan yang pada akhirnya akan mengarah ke nilai yang lebih baik. Format DTB yang sesuai dengan kebutuhan anak disleksia adalah Tobi DAISY, yang merupakan buku bicara dalam bentuk digital yang disinkronisasikan antara visual (teks, gambar, tabel, denah) dan audio. Hal ini dimungkinkan lebih memudahkan peserta didik disleksia dalam belajar membaca atau memahami bacaan. Tujuan kajian artikel ini adalah memberikan gambaran pengembangan Tobi DAISY yang dapat digunakan dan diproduksi secara pribadi bagi pend erita disleksia sehingga sesuai dengan konten yang dibutuhkan. Melalui kelebihan yang dimiliki Tobi DAISY, maka dapat disimpulkan relevan untuk digunakan bagi penderita disleksia dalam membantu belajar membaca dan memahami bacaan.

Sign in / Sign up

Export Citation Format

Share Document