{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/crnn-a-joint-neural-network-for-redundancy","title":"CRNN: A Joint Neural Network for Redundancy Detection","arxiv_id":"1706.01069","date":"2017-06-04","proceeding":null,"authors":["Xinyu Fu","Eugene Ch'ng","Uwe Aickelin","Simon See"],"abstract":"This paper proposes a novel framework for detecting redundancy in supervised\nsentence categorisation. Unlike traditional singleton neural network, our model\nincorporates character-aware convolutional neural network (Char-CNN) with\ncharacter-aware recurrent neural network (Char-RNN) to form a convolutional\nrecurrent neural network (CRNN). Our model benefits from Char-CNN in that only\nsalient features are selected and fed into the integrated Char-RNN. Char-RNN\neffectively learns long sequence semantics via sophisticated update mechanism.\nWe compare our framework against the state-of-the-art text classification\nalgorithms on four popular benchmarking corpus. For instance, our model\nachieves competing precision rate, recall ratio, and F1 score on the\nGoogle-news data-set. For twenty-news-groups data stream, our algorithm obtains\nthe optimum on precision rate, recall ratio, and F1 score. For Brown Corpus,\nour framework obtains the best F1 score and almost equivalent precision rate\nand recall ratio over the top competitor. For the question classification\ncollection, CRNN produces the optimal recall rate and F1 score and comparable\nprecision rate. We also analyse three different RNN hidden recurrent cells'\nimpact on performance and their runtime efficiency. We observe that MGU\nachieves the optimal runtime and comparable performance against GRU and LSTM.\nFor TFIDF based algorithms, we experiment with word2vec, GloVe, and sent2vec\nembeddings and report their performance differences.","url_abs":"http://arxiv.org/abs/1706.01069v1","url_pdf":"http://arxiv.org/pdf/1706.01069v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"crnn-a-joint-neural-network-for-redundancy","repo_url":"https://github.com/2023-MindSpore-1/ms-code-132","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[{"task_slug":"benchmarking","task_name":"Benchmarking"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"sentence","task_name":"Sentence"},{"task_slug":"text-classification","task_name":"Text Classification"},{"task_slug":"text-classification-1","task_name":"text-classification"}],"methods":[{"method_slug":"gru","method_name":"GRU"},{"method_slug":"glove","method_name":"GloVe"},{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}