Learning task specific distributed paragraph representations using a 2-tier convolutional neural network

We introduce a type of 2-tier convolutional neural network model for learning distributed paragraph representations for a special task (e.g. paragraph or short document level sentiment analysis and text topic categorization). We decompose the paragraph semantics into 3 cascaded constitutes: word representation, sentence composition and document composition. Specifically, we learn distributed word representations by a continuous bag-of-words model from a large unstructured text corpus. Then, using these word representations as pre-trained vectors, distributed task specific sentence representations are learned from a sentence level corpus with task-specific labels by the first tier of our model. Using these sentence representations as distributed paragraph representation vectors, distributed paragraph representations are learned from a paragraph-level corpus by the second tier of our model. It is evaluated on DBpedia ontology classification dataset and Amazon review dataset. Empirical results show the effectiveness of our proposed learning model for generating distributed paragraph representations.

Learning higher-level features with convolutional restricted Boltzmann machines for sentiment analysis

In recent years, learning word vector representations has attracted much interest in Natural Language Processing. Word representations or embeddings learned using unsupervised methods help addressing the problem of traditional bag-of-word approaches which fail to capture contextual semantics. In this paper we go beyond the vector representations at the word level and propose a novel framework that learns higher-level feature representations of n-grams, phrases and sentences using a deep neural network built from stacked Convolutional Restricted Boltzmann Machines (CRBMs). These representations have been shown to map syntactically and semantically related n-grams to closeby locations in the hidden feature space. We have experimented to additionally incorporate these higher-level features into supervised classifier training for two sentiment analysis tasks: subjectivity classification and sentiment classification. Our results have demonstrated the success of our proposed framework with 4% improvement in accuracy observed for subjectivity classification and improved the results achieved for sentiment classification over models trained without our higher level features.

Extracting topical phrases from clinical documents

In clinical documents, medical terms are often expressed in multi-word phrases. Traditional topic modelling approaches relying on the “bag-of-words” assumption are not effective in extracting topic themes from clinical documents. This paper proposes to first extract medical phrases using an off-the-shelf tool for medical concept mention extraction, and then train a topic model which takes a hierarchy of Pitman-Yor processes as prior for modelling the generation of phrases of arbitrary length. Experimental results on patients’ discharge summaries show that the proposed approach outperforms the state-of-the-art topical phrase extraction model on both perplexity and topic coherence measure and finds more interpretable topics.

Mechanism of action of the tetraflex accommodative intraocular lens

PURPOSE:To investigate the mechanism of action of the Tetraflex (Lenstec Kellen KH-3500) accommodative intraocular lens (IOL). METHODS:Thirteen eyes of eight patients implanted with the Tetraflex accommodating IOL for at least 2 years underwent assessment of their objective amplitude-of-accommodation by autorefraction, anterior chamber depth and pupil size with optical coherence tomography, and IOL flexure with aberrometry, each viewing a target at 0.0 to 4.00 diopters of accommodative demand. RESULTS:Pupil size decreased by 0.62+/-0.41 mm on increasing accommodative demand, but the Tetraflex IOL was relatively fixed in position within the eye. The ocular aberrations of the eye changed with increased accommodative demand, but not in a consistent manner among individuals. Those aberrations that appeared to be most affected were defocus, vertical primary and secondary astigmatism, vertical coma, horizontal and vertical primary and secondary trefoil, and spherical aberration. CONCLUSIONS:Some of the reported near vision benefits of the Tetraflex accommodating IOL appear to be due to changes in the optical aberrations because of the flexure of the IOL on accommodative effort rather than forward movement within the capsular bag.

An assessment of the accuracy of electrolyte concentrations in neonatal parental nutrition

Focal points Over a six-week period in January and February 2002, 2ml samples were removed from all neonatal PN bags dispensed Samples were submitted for analysis of sodium, potassium and magnesium in triplicate by the hospital's clinical chemistry department using a Vitros Codac 950AT, dry slide, automated analyser Only 19.3, 7.1 and 30.4 per cent of measured sodium, potassium and magnesium concentrations respectively deviated by £5 per cent from stated bag concentrations The results indicate that it is possible that some electrolyte concentrations included in neonatal PN vary significantly from stated values