Municipal meeting minutes are formal records documenting the discussions and decisions of local government, yet their content is often lengthy, dense, and difficult for citizens to navigate. Automatic summarization can help address this challenge by producing concise summaries for each discussion subject. Despite its potential, research on summarizing discussion subjects in municipal meeting minutes remains largely unexplored, especially in low-resource languages, where the inherent complexity of these documents adds further challenges. A major bottleneck is the scarcity of datasets containing high-quality, manually crafted summaries, which limits the development and evaluation of effective summarization models for this domain. In this paper, we present CitiLink-Summ, a new corpus of European Portuguese municipal meeting minutes, comprising 120 documents and 2,880 manually hand-written summaries, each corresponding to a distinct discussion subject. Leveraging this dataset, we establish baseline results for automatic summarization in this domain, employing state-of-the-art generative models (e.g., BART, PRIMERA) as well as large language models (LLMs), evaluated with both lexical and semantic metrics such as ROUGE, BLEU, METEOR, and BERTScore. CitiLink-Summ provides the first benchmark for municipal-domain summarization in European Portuguese, offering a valuable resource for advancing NLP research on complex administrative texts.

The expansion of social networks has contributed to the propagation of information relevant to general audiences. However, this is small percentage compared to all the data shared in such online platforms, which also includes private/personal information, simple chat messages and the recent called 'fake news'. In this paper, we make an exploratory analysis on two social networks to extract features that are indicators of relevant information in social network messages. Our goal is to build accurate machine learning models that are capable of detecting what is journalistically relevant. We conducted two experiments on CrowdFlower to build a solid ground truth for the models, by comparing the number of evaluations per post against the number of posts classified. The results show evidence that increasing the number of samples will result in a better performance on the relevancy classification task, even when relaxing in the number of evaluations per post. In addition, results show that there are significant correlations between the relevance of a post and its interest and whether is meaningfully for the majority of people. Finally, we achieve approximately 80% accuracy in the task of relevance detection using a small set of learning algorithms.