3. Producción

Browse

Settings

Language: search.filters.language.SpanishSubject: search.filters.subject.Datasets

Search Results

Now showing 1 - 2 of 2
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Evaluation on embeddings application for Spanish automatic text clustering
    (Universidad de Tarapaca, 2024-01-01)
    The vast amount of information on the Internet, primarily composed of texts, makes clustering reliable information a complicated task. This research aims to improve the automatic clustering of Spanish texts by applying embeddings and unsupervised learning algorithms. Five datasets were used, and embedding generation techniques such as Word2Vec, FastText, Glove, BERT, and GPT-2 were applied. Models like K-means, HDBSCAN, and AutoEncoder combined with K-means were employed for clustering. The results showed that the AutoEncoder model combined with the K-means using Glove embeddings achieved superior performance with an accuracy of 0.92, NMI of 0.79, and ARI of 0.81 on the BBC News dataset. In other datasets, the results varied, but the AutoEncoder with the K-means model consistently outperformed other methods. We conclude that neural network models with AutoEncoder and K-means layer are highly effective for automatically clustering Spanish texts, especially when using high-quality embeddings like Glove.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Aurora Cáceres’ personal album in datasets: A theoretical-critical and methodological approach to analyze the sociocultural practices of Latin American women intellectuals between centuries (1880–1925) from the digital humanities
    (Academia Peruana de la Lengua, 2024-07-01)
    We propose an approach to the personal album of Aurora Cáceres (1872-1958) from the digital humanities and information sciences. Through its conversion into datasets, we seek to make this complex and heterogeneous document computationally readable and to critically review it, pondering the networked contents that allow dialogues for new horizons of meaning, without ignoring the semantic and ideological value of the material supports. We hope to overcome the problem of reducing these devices to sources for the extraction of isolated data limited to intimate or depoliticized spheres. With the album, Cáceres, like other intellectuals of the inter-century period, evidenced her intervention in the public sphere and her strategies to construct a history and a memory in which she, as a subject and not as an object, plays a leading role. This work represents a methodological contribution to study this kind of sources, whose heterogeneity also implies their subordination or exclusion, given the challenges of reading, classifying and preserving them.