It is shown that the usage of CRF weights in BERT-based architectures achieves noteworthy improvements in the overall NER task by approximately 12 %, and that in few-shot learning set-ups the effectiveness of CRF weights is much higher in smaller training sets.
With the rapid growth of digital text data, automatically extracting meaningful information from unstructured text has become essential for applications such as information retrieval, knowledge extraction, and question answering. Named Entity Recognition (NER) is a key task in Natural Language Processing (NLP) that identifies and classifies entities such as persons, organizations, locations, dates, and monetary values into predefined categories. A large portion of real-world data is unstructured, making NER essential for extracting structured information. This study evaluates two widely used Python-based NLP libraries, spaCy and NLTK, on a benchmark dataset (CoNLL-2003) and a custom news dataset from Kaggle. Performance was measured using precision, recall, and F1-score. Results show that spaCy outperforms NLTK in both accuracy and efficiency, especially for real-world text data, demonstrating its suitability for robust and scalable NER applications. BERT-based and transformer-based models are discussed in the related work section to provide insight into recent advancements in Named Entity Recognition techniques.
Ilakkiya.B, Hemanandhini.I.G· 2026 4th International Confe...· 0 citations
This research focuses on enterprise profiling in scenarios where large volumes of diverse texts—such as registration records, annual reports, news articles, and bidding notices—are continuously generated. Instead of relying solely on a single data representation or classification model, we developed a comprehensive natural language processing (NLP) pipeline for extracting key information and identifying industries. The pipeline consists of several steps. First, we use a BERT-BiLSTM-CRF model to identify core e/nterprise entities. Then, we combine TF-IDF with BERT embeddings to create a hybrid feature scheme that captures both lexical cues and contextual semantics. To address the challenge of imbalanced industry labels, we apply SMOTE in the dense semantic space and pair it with Focal Loss to enhance learning for minority classes. Additionally, we introduce a Stacking strategy to integrate outputs from different models, making predictions more stable. Tests on a self-compiled dataset covering ten national economic sectors and about 50,000 enterprises show that our method achieves a macro-F1 score of 95.4%. It outperforms traditional machine learning baselines and single deep learning models, offering more reliable recognition for minority classes. These results suggest that our framework is well-suited for applications such as supply chain partner discovery, industrial mapping, and targeted investment promotion.
Xin-Yi Xu· Applied and Computational En...· 0 citations
Named Entity Recognition (NER) is a natural language processing task concerned with identifying mentions of named entities and classifying them according to a predefined set of categories. Despite the success of NER in domains, where such data is abundant it remains a formidable challenge for low-resource languages such as Kannada. In this paper we discuss the possible ways to approach NER for the Kannada language.
We explore various research directions including rule-based methods statistical machine learning neural networks and transformers based tagging methodologies. We highlight the various challenges in achieving NER for such a language and propose a transformer based contextual tagging framework for labelling sequences.
We propose to use mBERT IndicBERT and XLM-RoBERTa language models pretrained on target and other related Indic language corpora and further fine-tune these models for the NER task. We discuss various aspects for experimentation including data collection labelling data preparation methods data-splits evaluation metrics comparison with other models hyper parameter tuning entity-wise analysis and error analysis.
P. M., H. G, S. N· International journal of res...· 0 citations
The lack of high-quality ground truth datasets to train machine learning (ML) models impedes the potential of artificial intelligence for science research. Scientific information extraction (SIE) from the literature using LLMs is emerging as a powerful approach to automate the creation of these datasets. However, existing LLM-based approaches and benchmarking studies for SIE focus on broad topics such as biomedicine and chemistry, are limited to choice-based tasks, and focus on extracting information from short and well-formatted text. The potential of SIE methods in complex, open-ended tasks is considerably under-explored. In this study, we use a domain that has been virtually ignored in SIE, namely virology, to address these research gaps. We design a unique, open-ended SIE task of extracting mutations in a given virus that modify its interaction with the host. We develop a new, multi-step retrieval augmented generation (RAG) framework called VILLA for SIE. In parallel, we curate a novel dataset of 629 mutations in ten influenza A virus proteins obtained from 293 scientific publications to serve as ground truth for the mutation extraction task. We demonstrate VILLA's superior performance using a novel and comprehensive evaluation and comparison with vanilla RAG and other state-of-the art RAG- and agent-based tools for SIE. Finally, we evaluate the generalizability of our proposed method using another unique dataset of mutations in hepatitis E virus.
Blessy Antony, Amartya Dutta, Sneha Aggarwal et al.· Proceedings of the 32nd ACM...· 0 citations
Building upon recent Deep Neural Network architectures, current approaches lying in the intersection of Computer Vision and Natural Language Processing have achieved unprecedented breakthroughs in tasks like automatic captioning or image retrieval. Most of these learning methods, though, rely on large training sets of images associated with human annotations that specifically describe the visual content. In this paper we propose to go a step further and explore the more complex cases where textual descriptions are loosely related to the images. We focus on the particular domain of news articles in which the textual content often expresses connotative and ambiguous relations that are only suggested but not directly inferred from images. We introduce an adaptive CNN architecture that shares most of the structure for multiple tasks including source detection, article illustration and geolocation of articles. Deep Canonical Correlation Analysis is deployed for article illustration, and a new loss function based on Great Circle Distance is proposed for geolocation. Furthermore, we present BreakingNews, a novel dataset with approximately 100K news articles including images, text and captions, and enriched with heterogeneous meta-data (such as GPS coordinates and user comments). We show this dataset to be appropriate to explore all aforementioned problems, for which we provide a baseline performance using various Deep Learning architectures, and different representations of the textual and visual features. We report very promising results and bring to light several limitations of current state-of-the-art in this kind of domain, which we hope will help spur progress in the field.
Unknown authors· Journal of Science & Technol...· 0 citations
ReTaT is a dataset that can be used to train and evaluate systems for extracting relations whose expression spans the two modalities and its quality and suitability were assessed for the joint table-text relation extraction task using Large Language Models (LLMs).
Mohamed Ettaleb, Thibault Ehrhart, Nathalie Aussenac-Gilles et al.· International Conference on...· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
