Christophe Van Gysel

Unsupervised, Efficient and Semantic Expertise Retrieval

We introduce an unsupervised discriminative model for the task of retrieving experts in online document collections. We exclusively employ textual evidence and avoid explicit feature engineering by learning distributed word representations in an unsupervised way. We compare our model to state-of-the-art unsupervised statistical vector space and probabilistic generative approaches. Our proposed log-linear model achieves the retrieval performance levels of state-of-the-art document-centric methods with the low inference cost of so-called profile-centric approaches. It yields a statistically significant improved ranking over vector space and generative models in most cases, matching the performance of supervised methods on various benchmarks. That is, by using solely text we can do as well as methods that work with external evidence and/or relevance feedback. A contrastive analysis of rankings produced by discriminative and generative approaches shows that they have complementary strengths due to the ability of the unsupervised discriminative model to perform semantic matching.

Learning Latent Vector Spaces for Product Search

We introduce a novel latent vector space model that jointly learns the latent representations of words, e-commerce products and a mapping between the two without the need for explicit annotations. The power of the model lies in its ability to directly model the discriminative relation between products and a particular word. We compare our method to existing latent vector space models (LSI, LDA and word2vec) and evaluate it as a feature in a learning to rank setting. Our latent vector space model achieves its enhanced performance as it learns better product representations. Furthermore, the mapping from words to products and the representations of words benefit directly from the errors propagated back from the product representations during parameter estimation. We provide an in-depth analysis of the performance of our model and analyze the structure of the learned representations.

Pyndri: a Python Interface to the Indri Search Engine

We introduce pyndri, a Python interface to the Indri search engine. Pyndri allows to access Indri indexes from Python at two levels: (1) dictionary and tokenized document collection, (2) evaluating queries on the index. We hope that with the release of pyndri, we will stimulate reproducible, open and fast-paced IR research.

Reply With: Proactive Recommendation of Email Attachments

Email responses often contain items---such as a file or a hyperlink to an external document---that are attached to or included inline in the body of the message. Analysis of an enterprise email corpus reveals that 35% of the time when users include these items as part of their response, the attachable item is already present in their inbox or sent folder. A modern email client can proactively retrieve relevant attachable items from the users past emails based on the context of the current conversation, and recommend them for inclusion, to reduce the time and effort involved in composing the response. In this paper, we propose a weakly supervised learning framework for recommending attachable items to the user. As email search systems are commonly available, we constrain the recommendation task to formulating effective search queries from the context of the conversations. The query is submitted to an existing IR system to retrieve relevant items for attachment. We also present a novel strategy for generating labels from an email corpus---without the need for manual annotations---that can be used to train and evaluate the query formulation model. In addition, we describe a deep convolutional neural network that demonstrates satisfactory performance on this query formulation task when evaluated on the publicly available Avocado dataset and a proprietary dataset of internal emails obtained through an employee participation program.

Neural Vector Spaces for Unsupervised Information Retrieval

We propose the Neural Vector Space Model (NVSM), a method that learns representations of documents in an unsupervised manner for news article retrieval. In the NVSM paradigm, we learn low-dimensional representations of words and documents from scratch using gradient descent and rank documents according to their similarity with query representations that are composed from word representations. We show that NVSM performs better at document ranking than existing latent semantic vector space methods. The addition of NVSM to a mixture of lexical language models and a state-of-the-art baseline vector space model yields a statistically significant increase in retrieval effectiveness. Consequently, NVSM adds a complementary relevance signal. Next to semantic matching, we find that NVSM performs well in cases where lexical matching is needed. NVSM learns a notion of term specificity directly from the document collection without feature engineering. We also show that NVSM learns regularities related to Luhn significance. Finally, we give advice on how to deploy NVSM in situations where model selection (e.g., cross-validation) is infeasible. We find that an unsupervised ensemble of multiple models trained with different hyperparameter values performs better than a single cross-validated model. Therefore, NVSM can safely be used for ranking documents without supervised relevance judgments.

Lexical Query Modeling in Session Search

Lexical query modeling has been the leading paradigm for session search. In this paper, we analyze TREC session query logs and compare the performance of different lexical matching approaches for session search. Naive methods based on term frequency weighing perform on par with specialized session models. In addition, we investigate the viability of lexical query models in the setting of session search. We give important insights into the potential and limitations of lexical query modeling for session search and propose future directions for the field of session search.

Semantic Entities

Entity retrieval has seen a lot of interest from the research community over the past decade. Ten years ago, the expertise retrieval task gained popularity in the research community during the TREC Enterprise Track [10]. It has remained relevant ever since, while broadening to social media, to tracking the dynamics of expertise [1-5, 8, 11], and, more generally, to a range of entity retrieval tasks. In the talk, which will be given by the second author, we will point out that existing methods to entity or expert retrieval fail to address key challenges: (1) Queries and expert documents use different representations to describe the same concepts [6, 7]. Term mismatches between entities and experts [7] occur due to the inability of widely used maximum-likelihood language models to make use of semantic similarities between words [9]. (2) As the amount of available data increases, the need for more powerful approaches with greater learning capabilities than smoothed maximum-likelihood language models is obvious [13]. (3) Supervised methods for entity or expertise retrieval [5, 8] were introduced at the turn of the last decade. However, the acceleration of data availability has the major disadvantage that, in the case of supervised methods, manual annotation efforts need to sustain a similar order of growth. This calls for the further development of unsupervised methods. (4) According to some entity or expertise retrieval methods, a language model is constructed for every document in the collection. These methods lack efficient query capabilities for large document collections, as each query term needs to be matched against every document [2]. In the talk we will discuss a recently proposed solution [12] that has a strong emphasis on unsupervised model construction, efficient query capabilities and, most importantly, semantic matching between query terms and candidate entities. We show that the proposed approach improves retrieval performance compared to generative language models mainly due to its ability to perform semantic matching [7]. The proposed method does not require any annotations or supervised relevance judgments and is able to learn from raw textual evidence and document-candidate associations alone. The purpose of the proposal is to provide insight in how we avoid explicit annotations and feature engineering and still obtain semantically meaningful retrieval results. In the talk we will provide a comparative error analysis between the proposed semantic entity retrieval model and traditional generative language models that perform exact matching, which yields important insights in the relative strengths of semantic matching and exact matching for the expert retrieval task in particular and entity retrieval in general. We will also discuss extensions of the proposed model that are meant to deal with scalability and dynamic aspects of entity and expert retrieval.

Pytrec_eval: An Extremely Fast Python Interface to trec_eval

We introduce pytrec_eval, a Python interface to the tree_eval information retrieval evaluation toolkit. pytrec_eval exposes the reference implementations of trec_eval within Python as a native extension. We show that pytrec_eval is around one order of magnitude faster than invoking trec_eval as a sub process from within Python. Compared to a native Python implementation of NDCG, pytrec_eval is twice as fast for practically-sized rankings. Finally, we demonstrate its effectiveness in an application where pytrec_eval is combined with Pyndri and the OpenAI Gym where query expansion is learned using Q-learning.

Structural Regularities in Text-based Entity Vector Spaces

Entity retrieval is the task of finding entities such as people or products in response to a query, based solely on the textual documents they are associated with. Recent semantic entity retrieval algorithms represent queries and experts in finite-dimensional vector spaces, where both are constructed from text sequences. We investigate entity vector spaces and the degree to which they capture structural regularities. Such vector spaces are constructed in an unsupervised manner without explicit information about structural aspects. For concreteness, we address these questions for a specific type of entity: experts in the context of expert finding. We discover how clusterings of experts correspond to committees in organizations, the ability of expert representations to encode the co-author graph, and the degree to which they encode academic rank. We compare latent, continuous representations created using methods based on distributional semantics (LSI), topic models (LDA) and neural networks (word2vec, doc2vec, SERT). Vector spaces created using neural methods, such as doc2vec and SERT, systematically perform better at clustering than LSI, LDA and word2vec. When it comes to encoding entity relations, SERT performs best.