Dataset of Brazilian Federal Senate Session Transcriptions From 2023 with Relevant Topics and Stance Detection Annotations.

<p><strong>[pt-BR] Conjunto de dados de transcri&ccedil;&otilde;es de sess&otilde;es do Senado Federal brasileiro de 2023 com anota&ccedil;&otilde;es de t&oacute;picos relevantes e de dete&ccedil;&atilde;o de posicionamento.</strong></p> <p><strong>Dataset description</strong></p> <p>This set contains transcript data from 203 Federal Senate sessions from the year 2023, with annotations of relevant topics and positioning detection.</p> <p>The file <strong><a href="../api/records/11106904/draft/files/meetings.csv/content" target="_blank" rel="noopener">meetings.csv</a></strong> has the following structure:</p> <p>session_id: Unique event identifier<br>speaker_name: Name of the person who gave the speech<br>party: Political party<br>speech: Speech given</p> <p>The folder <a href="../api/records/11106904/draft/files/ground_truth.zip/content" target="_blank" rel="noopener noreferrer"><strong>ground_truth.zip</strong></a> contains 6 JSON annotation files related to the detection of relevant topics and positions, and its main keys are the following:</p> <p>id_session: Unique event identifier</p> <p>response: Model response<br>&nbsp; &nbsp; &nbsp;list_latent_topics: List of latent topics considered by the model<br>&nbsp; &nbsp; &nbsp;stances: The stances of each person, according to the model, on a given topic<br>model_response_evaluation: Evaluation of the model's response according to a human annotation<br>&nbsp; &nbsp; mapping: Mapping between the topic named by the model and the topic named in the annotation<br>&nbsp; &nbsp; list_latent_topics: It contains four keys which are the lists of topics considered true positives, false positives, true negatives and false negatives.</p> <p><strong>Code used</strong></p> <p>All the code used to process the data can be found at:</p> <p><a href="https://github.com/helenbc/tcc-notas-taquigraficas">https://github.com/helenbc/tcc-notas-taquigraficas</a></p> <p><strong>[pt-BR] Descri&ccedil;&atilde;o do conjunto de dados</strong></p> <p>Este conjunto cont&eacute;m dados de transcri&ccedil;&atilde;o de 203 sess&otilde;es do Senado Federal do ano de 2023, com anota&ccedil;&otilde;es de t&oacute;picos relevantes e detec&ccedil;&atilde;o de posicionamento.</p> <p>O arquivo meetings.csv possui a seguinte estrutura:</p> <p>session_id: Identificador &uacute;nico do evento&nbsp;</p> <p>speaker_name: Nome da pessoa que fez o discurso&nbsp;</p> <p>party: Partido pol&iacute;tico</p> <p>speech: Discurso proferido</p> <p>A pasta ground_truth.zip cont&eacute;m 6 arquivos de anota&ccedil;&atilde;o JSON relacionados &agrave; detec&ccedil;&atilde;o de t&oacute;picos relevantes e posi&ccedil;&otilde;es, e suas principais chaves s&atilde;o as seguintes:</p> <p>id_session: Identificador &uacute;nico do evento</p> <p>response: Resposta do modelo</p> <p>&nbsp; &nbsp; list_latent_topics: Lista de t&oacute;picos latentes considerados pelo modelo</p> <p>&nbsp; &nbsp; stances: As posi&ccedil;&otilde;es de cada pessoa, de acordo com o modelo, sobre um determinado t&oacute;pico</p> <p>model_response_evaluation: Avalia&ccedil;&atilde;o da resposta do modelo de acordo com uma anota&ccedil;&atilde;o humana</p> <p>&nbsp; &nbsp; mapping: Mapeamento entre o t&oacute;pico nomeado pelo modelo e o t&oacute;pico nomeado na anota&ccedil;&atilde;o</p> <p>&nbsp; &nbsp; list_latent_topics: Cont&eacute;m quatro chaves que s&atilde;o as listas de t&oacute;picos considerados verdadeiros positivos, falsos positivos, verdadeiros negativos e falsos negativos.</p> <p><strong>C&oacute;digo utilizado</strong></p> <p>Todo o c&oacute;digo utilizado para processar os dados pode ser encontrado em:</p> <p><a href="https://github.com/helenbc/tcc-notas-taquigraficas" target="_new" rel="noreferrer">https://github.com/helenbc/tcc-notas-taquigraficas</a></p> <p>&nbsp;</p> <p>&nbsp;</p>