The present work advances the accuracy and training speed of discriminative parsing. Our discriminative parsing method has no generative component, yet surpasses a generative baseline on constituent parsing, and does so with minimal linguistic cleverness. Our model can incorporate arbitrary features of the input and parse state, and performs feature selection incrementally over an exponential f...

The neuroscientific literature provides evidence that meditation may have measurable effects on the electrophysiological activity of the brain. We utilize a single-sensor EEG device to compare meditation and baseline epochs from a cohort of 31 long-term meditators from Tibetan and Indian monastic backgrounds. We show that our support vector machine approach can distinguish between these two sta...

We propose Regularized Max Pooling (RMP) for image classification. RMP classifies an image (or an image region) by extracting feature vectors at multiple subwindows at multiple locations and scales. Unlike Spatial Pyramid Matching where the subwindows are defined purely based on geometric correspondence, RMP accounts for the deformation of discriminative parts. The amount of deformation and the...

This paper introduces the different systems developed by Aholab Signal Processing Laboratory for The INTERSPEECH 2017 Computational Paralinguistics Challenge, which includes three different subtasks: Addressee, Cold and Snoring classification. Several classification strategies and features related with the spectrum, prosody and phase have been tested separately and further combined by using dif...

Much of the information used by an organization is collected in the form of manuals, regulations, news etc. These are grouped into controlled documentary collections, which are normally digitized and accessible via a content management system. However, obtaining new knowledge from collected documents in an organization requires not only sound search and retrieval of information tools, but also ...

We present a novel learning method for word embeddings designed for relation classification. Our word embeddings are trained by predicting words between noun pairs using lexical relation-specific features on a large unlabeled corpus. This allows us to explicitly incorporate relationspecific information into the word embeddings. The learned word embeddings are then used to construct feature vect...

Chinese event descriptive clause splitting is the task of splitting a complex Chinese sentence into several clauses. In this paper, we present a discriminative approach for Chinese event descriptive clause splitting task. By formulating the Chinese clause splitting task as a sequence labeling problem, we apply the structured SVMs model to Chinese clause splitting. Compared with other two baseli...

background and aims: since the produced recombinant proteins by molecular genetics techniques commonly have some limitations in the application, chimeric protein are introduced. chimeric proteins have found widespread application for the study of protein folding, structure stability, function and immunogenisity. methods: according to the known immunomodulatory effect and structure of hsp70 mole...

The ImageCLEF 2006 medical automatic annotation task encompasses 11,000 images from 116 categories, compared to 57 categories for 10,000 images of the similar task in 2005. As a baseline for comparison, a run using the same classifiers with the identical parameterization as in 2005 is submitted. In addition, the parameterization of the classifier was optimized according to the 9,000/1,000 split...

Learning to predict signiicant events from sequences of data with categorical features is an important problem in many application areas. We focus on events for system management, and formulate the problem of prediction as a classiication problem. We perform co-occurrence analysis of events by means of Singular Value Decomposition (SVD) of the examples constructed from the data. This process is...