Faegheh Hasibi

Entity Linking in Queries: Tasks and Evaluation

Annotating queries with entities is one of the core problem areas in query understanding. While seeming similar, the task of entity linking in queries is different from entity linking in documents and requires a methodological departure due to the inherent ambiguity of queries. We differentiate between two specific tasks, semantic mapping and interpretation finding, discuss current evaluation methodology, and propose refinements. We examine publicly available datasets for these tasks and introduce a new manually curated dataset for interpretation finding. To further deepen the understanding of task differences, we present a set of approaches for effectively addressing these tasks and report on experimental results.

Exploiting Entity Linking in Queries for Entity Retrieval

The premise of entity retrieval is to better answer search queries by returning specific entities instead of documents. Many queries mention particular entities; recognizing and linking them to the corresponding entry in a knowledge base is known as the task of entity linking in queries. In this paper we make a first attempt at bringing together these two, i.e., leveraging entity annotations of queries in the entity retrieval model. We introduce a new probabilistic component and show how it can be applied on top of any term-based entity retrieval model that can be emulated in the Markov Random Field framework, including language models, sequential dependence models, as well as their fielded variations. Using a standard entity retrieval test collection, we show that our extension brings consistent improvements over all baseline methods, including the current state-of-the-art. We further show that our extension is robust against parameter settings.

On the Reproducibility of the TAGME Entity Linking System

Reproducibility is a fundamental requirement of scientific research. In this paper, we examine the repeatability, reproducibility, and generalizability of TAGME, one of the most popular entity linking systems. By comparing results obtained from its public API with (re)implementations from scratch, we obtain the following findings. The results reported in the TAGME paper cannot be repeated due to the unavailability of data sources. Part of the results are reproducible through the provided API, while the rest are not reproducible. We further show that the TAGME approach is generalizable to the task of entity linking in queries. Finally, we provide insights gained during this process and formulate lessons learned to inform future reducibility efforts.

DBpedia-Entity v2: A Test Collection for Entity Search

The DBpedia-entity collection has been used as a standard test collection for entity search in recent years. We develop and release a new version of this test collection, DBpedia-Entity v2, which uses a more recent DBpedia dump and a unified candidate result pool from the same set of retrieval models. Relevance judgments are also collected in a uniform way, using the same group of crowdsourcing workers, following the same assessment guidelines. The result is an up-to-date and consistent test collection.To facilitate further research, we also provide details about the pre-processing and indexing steps, and include baseline results from both classical and recently developed entity search methods.

Target Type Identification for Entity-Bearing Queries

Identifying the target types of entity-bearing queries can help improve retrieval performance as well as the overall search experience. In this work, we address the problem of automatically detecting the target types of a query with respect to a type taxonomy. We propose a supervised learning approach with a rich variety of features. Using a purpose-built test collection, we show that our approach outperforms existing methods by a remarkable margin.

Nordlys: A Toolkit for Entity-Oriented and Semantic Search

We introduce Nordlys, a toolkit for entity-oriented and semantic search. It provides functionality for entity cataloging, entity retrieval, entity linking, and target type identification. Nordlys may be used as a Python library or as a RESTful API, and also comes with a web-based user interface. The toolkit is open source and is available at http://nordlys.cc.

Entity Linking in Queries: Efficiency vs. Effectiveness

Identifying and disambiguating entity references in queries is one of the core enabling components for semantic search. While there is a large body of work on entity linking in documents, entity linking in queries poses new challenges due to the limited context the query provides coupled with the efficiency requirements of an online setting. Our goal is to gain a deeper understanding of how to approach entity linking in queries, with a special focus on how to strike a balance between effectiveness and efficiency. We divide the task of entity linking in queries to two main steps: candidate entity ranking and disambiguation, and explore both unsupervised and supervised alternatives for each step. Our main finding is that best overall performance (in terms of efficiency and effectiveness) can be achieved by employing supervised learning for the entity ranking step, while tackling disambiguation with a simple unsupervised algorithm. Using the Entity Recognition and Disambiguation Challenge platform, we further demonstrate that our recommended method achieves state-of-the-art performance.

A greedy algorithm for finding sets of entity linking interpretations in queries

We describe our participation in the short text track of the Entity Recognition and Disambiguation (ERD) challenge, where the task is to find all interpretations of entity-related queries and link them to entities in a knowledge base. We approached this task using a multi-stage framework. First, we recognize entity mentions based on known surface forms. Next, we score candidate entities using a learning-to-rank method. Finally, we use a greedy algorithm to find all valid interpretation sets for the query. We report on evaluation results using the official ERD challenge platform.

Towards a Unified Supervised Approach for Ranking Triples of Type-Like Relations

Knowledge bases play a crucial role in modern search engines and provide users with information about entities. A knowledge base may contain many facts (i.e., RDF triples) about an entity, but only a handful of them are of significance for a searcher. Identifying and ranking these RDF triples is essential for various applications of search engines, such as entity ranking and summarization. In this paper, we present the first effort towards a unified supervised approach to rank triples from various type-like relations in knowledge bases. We evaluate our approach using the recently released test collections from the WSDM Cup 2017 and demonstrate the effectiveness of the proposed approach despite the fact that no relation-specific feature is used.