• Corpus ID: 27163178

Overview of NTCIR-13 Actionable Knowledge Graph (AKG) Task

@inproceedings{Blanco2017OverviewON,
  title={Overview of NTCIR-13 Actionable Knowledge Graph (AKG) Task},
  author={Roi Blanco and Hideo Joho and Adam Jatowt and Haitao Yu and Shuhei Yamamoto},
  booktitle={NTCIR},
  year={2017}
}
This paper overviews NTCIR-13 Actionable Knowledge Graph (AKG) task. The task focuses on finding possible actions related to input entities and the relevant properties of such actions. AKG is composed of two subtasks: Action Mining (AM) and Actionable Knowledge Graph Generation (AKGG). Both subtasks are focused on English language. 9 runs have been submitted by 4 teams for the task. In this paper we describe both the subtasks, datasets, evaluation methods and the results of meta analyses. 

Figures and Tables from this paper

TUA 1 at the NTCIR-13 Actionable Knowledge Graph Task : Sampling Related Actions from Online Searching

TLDR
This work focuses on sequentially sampling the most related actions for any named entity based on online search results, and proposes three criteria, i.e. significance, representativeness, and diverseness, for evaluating the relatedness of candidate actions in the search results.

CUIS Team for NTCIR-13 AKG Task

TLDR
This paper describes the approach for Actionable Knowledge Graph (AKG) task at NTCIR-13, and employs supervised learning technique to improve performance by minimizing a simple position-sensitive loss function on additional manually annotated training data from the dry run topics.

Test Collection for Evaluating Actionable Knowledge Graphs

TLDR
The NTCIR-13 Actionable Knowledge Graph (AKG) task and its test collections are discussed, which focuses on pos- sible actions related to input entities as well as the relevant properties of such actions.

TLAB at the NTCIR-13 AKG Task

TLDR
A probabilistic model is employed to address the Action Mining (AM) problem and is presented as an essential step and relatively new research direction to nurture research on generating such KG that is optimized for facilitating entity’s actions.

CrossBERT: A Triplet Neural Architecture for Ranking Entity Properties

TLDR
This work proposes a new method for property ranking, CrossBERT, which builds on the Bidirectional Encoder Representations from Transformers (BERT) and creates a new triplet network structure on cross query-property pairs that is used to rank properties.

Etna: Harvesting Action Graphs from Websites

TLDR
To harvest action knowledge from websites, the proposed Etna constructs an action-state model of the website visualized as an action graph, which includes definitions of tasks and actions, knowledge about their start/end states, and execution scripts for their automation.

Overview of NTCIR-13

TLDR
This paper describes an outline of the research project, which includes its organization, schedule, scope and task designs, and introduces brief statistics of participants in the NTCIR-13 Conference.

Query Understanding via Entity Attribute Identification

TLDR
This study introduces the task of entity attribute identification and proposes two methods to address it: a model based on Markov Random Field, and a learning to rank model.

Toward action comprehension for searching: Mining actionable intents in query entities

TLDR
This article presents a novel research method for mining the actionable intents for search users, by generating a ranked list of the potentially most informative actions based on a massive pool of action samples.

Knowledge Graphs: An Information Retrieval Perspective

TLDR
An overview of the literature on knowledge graphs (KGs) in the context of information retrieval (IR) is provided and how KGs can be employed to support IR tasks, including document and entity retrieval is discussed.

References

SHOWING 1-10 OF 12 REFERENCES

TUA 1 at the NTCIR-13 Actionable Knowledge Graph Task : Sampling Related Actions from Online Searching

TLDR
This work focuses on sequentially sampling the most related actions for any named entity based on online search results, and proposes three criteria, i.e. significance, representativeness, and diverseness, for evaluating the relatedness of candidate actions in the search results.

CUIS Team for NTCIR-13 AKG Task

TLDR
This paper describes the approach for Actionable Knowledge Graph (AKG) task at NTCIR-13, and employs supervised learning technique to improve performance by minimizing a simple position-sensitive loss function on additional manually annotated training data from the dry run topics.

Overview of NTCIR-13

TLDR
This paper describes an outline of the research project, which includes its organization, schedule, scope and task designs, and introduces brief statistics of participants in the NTCIR-13 Conference.

Lightweight Multilingual Entity Extraction and Linking

TLDR
An accurate and lightweight, multilingual named entity recognition (NER) and linking (NEL) system that achieves state-of-the-art performance on TAC KBP 2013 multilingual data and on English AIDA CONLL data is presented.

Active objects: actions for entity-centric search

TLDR
This paper introduces an entity-centric search experience, called Active Objects, in which entity-bearing queries are paired with actions that can be performed on entities, and proposes an evaluation methodology for measuring the relevance of the recommended actions.

Fast and Space-Efficient Entity Linking for Queries

TLDR
This paper proposes a probabilistic model that leverages user-generated information on the web to link queries to entities in a knowledge base and significantly outperforms several state-of-the-art baselines while being able to process queries in sub-millisecond times---at least two orders of magnitude faster than existing systems.

Role-explicit query identification and intent role annotation

TLDR
A simplified word n-gram role model (SWNR) is proposed, which quantifies the generating probability of a role-explicit query and performs intent role annotation effectively and builds classifiers to address RP-2 in a supervised manner.

A web of concepts

TLDR
The case for developing a web of concepts is made by starting with the current view of web, extracting concept-centric metadata, and stitching it together to create a semantically rich aggregate view of all the information available on the web for each concept instance.

A taxonomy of web search

TLDR
This taxonomy of web searches is explored and how global search engines evolved to deal with web-specific needs is discussed.

A Study of Smoothing Methods for Language Models Applied to Ad Hoc Information Retrieval

TLDR
This paper examines the sensitivity of retrieval performance to the smoothing parameters and compares several popular smoothing methods on different test collection.