What Does a TextCNN Learn?
نویسندگان
چکیده
TextCNN, the convolutional neural network for text, is a useful deep learning algorithm for sentence classification tasks such as sentiment analysis and question classification[2]. However, neural networks have long been known as black boxes because interpreting them is a challenging task. Researchers have developed several tools to understand a CNN for image classification by deep visualization[6], but research about deep TextCNNs is still insufficient. In this paper, we are trying to understand what a TextCNN learns on two classical NLP datasets. Our work focuses on functions of different convolutional kernels and correlations between convolutional kernels.
منابع مشابه
Adopting New International Health Instruments – What Can We Learn From the FCTC?; Comment on “The Legal Strength of International Health Instruments - What It Brings to Global Health Governance?”
This Commentary forms a response to Nikogosian’s and Kickbusch’s forward-looking perspective about the legal strength of international health instruments. Building on their arguments, in this commentary we consider what we can learn from the Framework Convention on Tobacco Control (FCTC) for the adoption of new legal international health instruments.
متن کاملHow Can a Global Social Support System Hope to Achieve Fairer Competiveness?; Comment on “A Global Social Support System: What the International Community Could Learn From the United States’ National Basketball Association”
Ooms et al sets out some good general principles for a global social support system to improve fairer global competitiveness as a result of redistribution. This commentary sets out to summarize some of the conditions that would need to be satisfied for it to level up gradients in inequality through such a social support system, using the National Basketball Association (NBA) example as a point ...
متن کاملOne Semester of Occupational Therapy Passed Virtually: What Experiences Did We Learn from Covid-19?
One Semester of Occupational Therapy Passed Virtually: What Experiences Did We Learn from Covid-19? Masoud Gharib 1, Seyed Ali Hosseini 2 1 Assistant Professor, Department of Occupational, Orthopedic Research Center, Mazandaran University of Medical Sciences, Sari, Iran 2 Professor, Department of Occupational Therapy, University of Social Welfare and Rehabilitation Sciences, Tehran...
متن کاملMistaking the Map for the Territory: What Society Does With Medicine; Comment on “Medicalisation and Overdiagnosis: What Society Does to Medicine”
Van Dijk et al describe how society’s influence on medicine drives both medicalisation and overdiagnosis, and allege that a major political and ethical concern regarding our increasingly interpreting the world through a biomedical lens is that it serves to individualise and depoliticize social problems. I argue that for medicalisation to serve this purpose, it would have to exclude the possibil...
متن کاملCOVID-19 Pandemic: What Can the West Learn From the East?
Differences in public health approaches to control the coronavirus disease 2019 (COVID-19) pandemic could largely explain substantial variations in epidemiological indicators (such as incidence and mortality) between the West and the East. COVID-19 revealed vulnerabilities of most western countries’ healthcare systems in their response to the ongoing public health crisi...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
- CoRR
دوره abs/1801.06287 شماره
صفحات -
تاریخ انتشار 2018