The paper details a scheme for lossless compression of short data series larger than 50 Bytes. The method uses arithmetic coding and context modeling with a low-complexity data model. A data model that takes 32 kBytes of RAM already cuts the data size in half. The compression scheme just takes a few pages of source code, is scalable in memory size, and may be useful in sensor or cellular networ...

This paper presents corpus-based methods to find similarity between short text (sentences, paragraphs, ...) which has many applications in the field of NLP. Previous works on this problem have been based on supervised methods or have used external resources such as WordNet, British National Corpus etc. Our methods are focused on unsupervised corpus-based methods. We present a new method, based ...

This paper analyzed three middle grade novels and a short story from two cultures using the journey motif as the vehicle for the analysis. The three novels are Bud Not Buddy and The Watsons Go to Birmingham both by Paul Curtis   and Journey to Jo’burg by Naidoo (1986). The short story is My Two Dads by M. Lee.  The three novels and a short story were chosen because these no...

Short text clustering has far-reaching effects on semantic analysis, showing its importance for multiple applications such as corpus summarization and information retrieval. However, it inevitably encounters the severe sparsity of short representations, making previous approaches still far from satisfactory. In this paper, we present a novel attentive representation learning model shot clusteri...

although filling the gap in reading comprehension gained momentum with the rise of the top-down approach, vygotsky’ concept of scaffolding and the dual code  theory provided a strong support for the use of paratext to enhance comprehension. scaffolding is dependent on other-regulation, one type of which is object-regulation. from this vantage-point, various types of paratext can function as sou...

The use of background knowledge is largely unexploited in text classification tasks. This paper explores word taxonomies as means for constructing new semantic features, which may improve the performance and robustness learned classifiers. We propose tax2vec, a parallel algorithm taxonomy-based demonstrate its on six short problems: prediction gender, personality type, age, news topics, drug si...