Word embeddings are increasingly used for the automatic detection of semantic change; yet, a robust evaluation and systematic comparison of the choices involved has been lacking. We propose a new evaluation framework for semantic change detection and find that (i) using the whole time series is preferable over only comparing between the first and last time points; (ii) independently trained and aligned embeddings perform better than continuously trained embeddings for long time periods; and (iii) that the reference point for comparison matters. We also present an analysis of the changes detected on a large Twitter dataset spanning 5.5 years.

Benchmarking state-of-the-art text classification and information extraction systems in multilingual, cross-lingual, few-shot, and zeroshot settings for socio-political event information collection is achieved in the scope of the shared task Socio-political and Crisis Events Detection at the workshop CASE @ ACL-IJCNLP 2021. Socio-political event data is utilized for national and international policyand decision-making. Therefore, the reliability and validity of such datasets are of utmost importance. We split the shared task into three parts to address the three aspects of data collection (Task 1), fine-grained semantic classification (Task 2), and evaluation (Task 3). Task 1, which is the focus of this report, is on multilingual protest news detection and comprises four subtasks that are document classification (subtask 1), sentence classification (subtask 2), event sentence coreference identification (subtask 3), and event extraction (subtask 4). All subtasks have English, Portuguese, and Spanish for both training and evaluation data. Data in Hindi language is available only for the evaluation of subtask 1. The majority of the submissions, which are 238 in total, are created using multi-and cross-lingual approaches. Best scores are between 77.27 and 84.55 F1-macro for subtask 1, between 85.32 and 88.61 F1macro for subtask 2, between 84.23 and 93.03 CoNLL 2012 average score for subtask 3, and between 66.20 and 78.11 F1-macro for subtask 4 in all evaluation settings. The performance of the best system for subtask 4 is above 66.20 F1 for all available languages. Although there is still a significant room for improvement in cross-lingual and zero-shot settings, the best submissions for each evaluation scenario yield remarkable results. Monolingual models outperformed the multilingual models in a few evaluation scenarios, in which there is relatively much training data.

During the Covid-19 pandemic, universities in the UK used social media to raise awareness and provide guidance and advice about the disease to students and staff. We explain why some universities used social media to communicate with stakeholders sooner than others. To do so, we identified the date of the first Covid-19 related tweet posted by each university in the country and used survival models to estimate the effect of university-specific characteristics on the timing of these messages. In order to confirm our results, we supplemented our analysis with a study of the introduction of coronavirus-related university webpages. We find that universities with large numbers of students are more likely to use social media and the web to speak about the pandemic sooner than institutions with fewer students. Universities with large financial resources are also more likely to tweet sooner, but they do not introduce Covid-19 webpages faster than other universities. We also find evidence of a strong process of emulation, whereby universities are more likely to post a coronavirus-related tweet or webpage if other universities have already done so.

No abstract

We analyse Recurrent Neural Networks (RNNs) to understand the significance of multiple LSTM layers. We argue that the Weighted Finite-state Automata (WFA) trained using a spectral learning algorithm are helpful to analyse RNNs. Our results suggest that multiple LSTM layers in RNNs help learning distributed hidden states, but have a smaller impact on the ability to learn long-term dependencies. The analysis is based on the empirical results, however relevant theory (whenever possible) was discussed to justify and support our conclusions.