Search

Popular word embedding methods such as GloVe and Word2Vec are related to the factorization of the pointwise mutual information (PMI) matrix. In this paper, we link correspondence analysis (CA) to the factorization of the PMI matrix. CA is a dimensionality reduction method that uses singular value decomposition (SVD), and we show that CA is mathematically close to the weighted factorization of the PMI matrix. In addition, we present variants of CA that turn out to be successful in the factorization of the word-context matrix, i.e. CA applied to a matrix where the entries undergo a square-root transformation (ROOT-CA) and a root-root transformation (ROOTROOT-CA). While this study focuses on traditional static word embedding methods, to extend the contribution of this paper, we also include a comparison of transformer-based encoder BERT, i.e. contextual word embedding, with these traditional methods. An empirical comparison among CA- and PMI-based methods as well as BERT shows that overall results of ROOT-CA and ROOTROOT-CA are slightly better than those of the PMI-based methods and are competitive with BERT.

Correspondence analysis (CA) is a popular technique to visualize the relationship between two categorical variables. CA uses the data from a two-way contingency table and is affected by the presence of outliers. The supplementary points method is a popular method to handle outliers. Its disadvantage is that the information from entire rows or columns is removed. However, outliers can be caused by cells only. In this paper, a reconstitution algorithm is introduced to cope with such cells. This algorithm can reduce the contribution of cells in CA instead of deleting entire rows or columns. Thus the remaining information in the row and column involved can be used in the analysis. The reconstitution algorithm is compared with two alternative methods for handling outliers, the supplementary points method and MacroPCA. It is shown that the proposed strategy works well.

Both latent semantic analysis (LSA) and correspondence analysis (CA) are dimensionality reduction techniques that use singular value decomposition (SVD) for information retrieval. Theoretically, the results of LSA display both the association between documents and terms, and marginal effects; in comparison, CA only focuses on the associations between documents and terms. Marginal effects are usually not relevant for information retrieval, and therefore, from a theoretical perspective CA is more suitable for information retrieval. In this paper, we empirically compare LSA and CA. The elements of the raw document-term matrix are weighted, and the weighting exponent of singular values is adjusted to improve the performance of LSA. We explore whether these two weightings also improve the performance of CA. In addition, we compare the optimal singular value weighting exponents for LSA and CA to identify what the initial dimensions in LSA correspond to. The results for four empirical datasets show that CA always performs better than LSA. Weighting the elements of the raw data matrix can improve CA; however, it is data dependent and the improvement is small. Adjusting the singular value weighting exponent usually improves the performance of CA; however, the extent of the improved performance depends on the dataset and number of dimensions. In general, CA needs a larger singular value weighting exponent than LSA to obtain the optimal performance. This indicates that CA emphasizes initial dimensions more than LSA, and thus, margins play an important role in the initial dimensions in LSA.

Multi-hop question answering (QA) requires reasoning over multiple documents to answer a complex question and provide interpretable supporting evidence. However, providing supporting evidence is not enough to demonstrate that a model has performed the desired reasoning to reach the correct answer. Most existing multi-hop QA methods fail to answer a large fraction of sub-questions, even if their parent questions are answered correctly. In this paper, we propose the Prompt-based Conservation Learning (PCL) framework for multi-hop QA, which acquires new knowledge from multi-hop QA tasks while conserving old knowledge learned on single-hop QA tasks, mitigating forgetting. Specifically, we first train a model on existing single-hop QA tasks, and then freeze this model and expand it by allocating additional sub-networks for the multi-hop QA task. Moreover, to condition pre-trained language models to stimulate the kind of reasoning required for specific multi-hop questions, we learn soft prompts for the novel sub-networks to perform type-specific reasoning. Experimental results on the HotpotQA benchmark show that PCL is competitive for multi-hop QA and retains good performance on the corresponding single-hop sub-questions, demonstrating the efficacy of PCL in mitigating knowledge loss by forgetting., Comment: Accepted to COLING 2022

Automated question quality rating (AQQR) aims to evaluate question quality through computational means, thereby addressing emerging challenges in online learnersourced question repositories. Existing methods for AQQR rely solely on explicitly-defined criteria such as readability and word count, while not fully utilising the power of state-of-the-art deep-learning techniques. We propose DeepQR, a novel neural-network model for AQQR that is trained using multiple-choice-question (MCQ) datasets collected from PeerWise, a widely-used learnersourcing platform. Along with designing DeepQR, we investigate models based on explicitly-defined features, or semantic features, or both. We also introduce a self-attention mechanism to capture semantic correlations between MCQ components, and a contrastive-learning approach to acquire question representations using quality ratings. Extensive experiments on datasets collected from eight university-level courses illustrate that DeepQR has superior performance over six comparative models., Comment: EAAI 22

Latent semantic analysis (LSA) and correspondence analysis (CA) are two techniques that use a singular value decomposition (SVD) for dimensionality reduction. LSA has been extensively used to obtain low-dimensional representations that capture relationships among documents and terms. In this article, we present a theoretical analysis and comparison of the two techniques in the context of document-term matrices. We show that CA has some attractive properties as compared to LSA, for instance that effects of margins, i.e. sums of row elements and column elements, arising from differing document-lengths and term-frequencies are effectively eliminated, so that the CA solution is optimally suited to focus on relationships among documents and terms. A unifying framework is proposed that includes both CA and LSA as special cases. We empirically compare CA to various LSA based methods on text categorization in English and authorship attribution on historical Dutch texts, and find that CA performs significantly better. We also apply CA to a long-standing question regarding the authorship of the Dutch national anthem Wilhelmus and provide further support that it can be attributed to the author Datheen, amongst several contenders.