id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_ndtqmncagrai7ec23g3kd4qk34	Stefanos Angelidis	Multiple Instance Learning Networks for Fine-Grained Sentiment Analysis	2018	17	.pdf	application/pdf	9179	1067	59	Angelidis, S & Lapata, M 2018, 'Multiple Instance Learning Networks for Fine-Grained Sentiment Analysis', learns to predict the sentiment of text segments, i.e. sentences or elementary discourse Instead of learning from individually labeled segments, our model only requires document-level supervision and learns to introspectively judge the sentiment of constituent segments. final sentiment prediction is produced using a softmax classifier and the model is trained via backpropagation using sentence-level sentiment labels. 2When applied to the YELP'13 and IMDB document classification datasets, the use of CNNs results in a relative performance decrease of < 2% compared Yang et al's model (2016). Given a review, we predict the polarity of every segment, allowing for the extraction of sentiment-heavy opinions. classifier would learn to predict the document's sentiment by directly conditioning on its segments' feature representations or their aggregate: Figure 6 illustrates the distribution of polarity scores produced by the two models on the Yelp'13 dataset (sentence segmentation).	./cache/work_ndtqmncagrai7ec23g3kd4qk34.pdf	./txt/work_ndtqmncagrai7ec23g3kd4qk34.txt