The University of Alicante at MultiLing 2015: approach, results and further insights

Abstract

In this paper we present the approach and results of our participation in the 2015 MultiLing Single-document Summarization task. Our approach is based on the Principal Component Analysis (PCA) technique enhanced with lexical-semantic knowledge. For testing our approach, different configurations were set up, thus generating different types of summaries (i.e., generic and topic-focused), as well as testing some language-specific resources on top of the language-independent basic PCA approach, submitting a total of 6 runs for each selected language (English, German, and Spanish). Our participation in MultiLing has been very positive, ranking at intermediate positions when compared to the other participant systems, showing that PCA is a good technique for generating language-independent summaries, but the addition of lexical-semantic knowledge may heavily depend on the size and quality of the resources available for each language.

A study of semantic augmentation of word embeddings for extractive summarization

“…For example, in (Dang and Luo, 2008), the authors utilize Word-Net synsets as a keyphrase ranking mechanism, based on candidate synset relevance to the text. Other approaches (Vicente et al, 2015) use semantic features from Wordnet and named entity extrac-tion, followed by a PCA-based post-processing step for dimensionality reduction. Wordnet is also utilized in (Li et al, 2017) where the authors use the resource for sentence similarity extraction, using synset similarity on the word level and treating the resulting scores as additional features for summarization and citation linkage.…”

“…Our approach bears some similarities with the work of (Vicente et al, 2015), extending the investigation to additional post-processing techniques to PCA, examining post-processing application strategies, and adopting deep neural word embeddings as the lexical representation, while grounding on a number of baselines. In the following section, we will describe our approach in detail, including text representation, semantic feature extraction, training and evaluation.…”

Towards Personalized Summaries in Spanish based on Learning Styles Theory

“…Thenceforth, several theories in linguistics and artificial intelligence have been proposed. Such as, superficial techniques [4,5,6], graph-based techniques [7,8,9,10], algebraic reduction techniques [11], statistics [12], etc. However, these researches are still improving.…”