The digitization of Spanish historical newspapers poses significant challenges due to low scan quality, typographical diversity, complex layouts and linguistic variation from contemporary Spanish. While advances in Optical Character Recognition (OCR) and layout-aware models offer promising results, their effectiveness strongly depends on the quality and consistency of the underlying training corpora. This work focuses on corpus construction and evaluation for historical document processing. Two experiments were conducted. In the first corpus los101 was used, a manually curated and structurally annotated subcorpus derived from historical Spanish newspapers, designed to ensure coherent ground truth under heterogeneous real-world conditions. This corpus enables systematic experimentation across OCR and document layout analysis tasks. In a second experimental phase, we apply an additional layout-focused corpus characterized by structural regularity and consistent page organization, allowing us to isolate the impact of layout homogeneity on segmentation performance. State-of-the-art OCR models and a layout detection model are evaluated as validation instruments to assess corpus adequacy rather than as primary contributions. Quantitative and qualitative analyses based on (1) relationship between annotation quality, (2) structural variability, and (3) model behavior, show that heterogeneous corpora challenge both transcription and segmentation stability, while layout-consistent data significantly improves structural detection reliability.

This paper presents Q-WordNet, a lexical resource consisting of WordNet senses automatically annotated by positive and negative polarity. Polarity classification amounts to decide whether a text (sense, sentence, etc.) may be associated to positive or negative connotations. Polarity classification is becoming important within the fields of Opinion Mining and Sentiment Analysis for determining opinions about commercial products, on companies reputation management, brand monitoring, or to track attitudes by mining online forums, blogs, etc. Inspired by work on classification of word senses by polarity (e.g., SentiWordNet), and taking WordNet as a starting point, we build Q-WordNet. Instead of applying external tools such as supervised classifiers to annotated WordNet synsets by polarity, we try to effectively maximize the linguistic information contained in WordNet, thereby taking advantage of the human effort put by lexicographers and annotators. The resulting resource is a subset of WordNet senses classified as positive or negative. In this approach, neutral polarity is seen as the absence of positive or negative polarity. The evaluation of Q-WordNet shows an improvement with respect to previous approaches. We believe that Q-WordNet can be used as a starting point for data-driven approaches in sentiment analysis.