Financialisation has become a new buzz word in social sciences, but, although some of the earliest usages concept can be found with economic historians, recent fad largely been ignored by history. This is true also for Nordic region. survey article highlights handful studies on financialisation countries general and within history, particular, but more importantly, it relates history long wave ...

Word embeddings are a powerful approach for unsupervised analysis of language. Recently, Rudolph et al. (2016) developed exponential family embeddings, which cast word embeddings in a probabilistic framework. Here, we develop dynamic embeddings, building on exponential family embeddings to capture how the meanings of words change over time. We use dynamic embeddings to analyze three large colle...

Thirty-four cases of asthma in children referred to outpatient clinics in Newcastle upon Tyne (16 cases) and London (18 cases) were reviewed. In both cities there was evidence of inappropriate diagnosis and treatment by general practitioners. One of the main factors seemed to be doctors' reluctance to use the word "asthma," even when a history of episodic wheezing strongly suggested the diagnos...

This paper presents a tunable index compression scheme for supporting time-travel phrase queries over large versioned corpora such as web archives. Support for phrase queries makes maintenance of word positions necessary, thus increasing the index size significantly. We propose to fuse the word positions in many neighboring versions of a document, and thus exploit the typically high level of re...

A new language model for speech recognition inspired by linguistic analysis is presented. The model develops hidden hierarchical structure incrementally and uses it to extract meaningful information from the word history — thus enabling the use of extended distance dependencies — in an attempt to complement the locality of currently used trigram models. The structured language model, its probab...

We use recurrent neural network (RNN) based language models to improve the BUT English meeting recognizer. On the baseline setup using the original language models we decrease word error rate (WER) more than 1% absolute by n-best list rescoring and language model adaptation. When n-gram language models are trained on the same moderately sized data set as the RNN models, improvements are higher ...

men was rarely used in ancient Egypt, not until much later in its history, but the name stuck and has remained in popular usage. The Egyptians would never have used the word mummy, it has been suggested3 that their word would have been Sahu, which means, an embalmed cadaver. Long after mummification was abandoned, extracts of mummies were used for all sorts of purposes. The Arab physician Avice...

A new language model for speech recognition inspired by linguistic analysis is presented. The model develops hidden hierarchical structure incrementally and uses it to extract meaningful information from the word history — thus enabling the use of extended distance dependencies — in an attempt to complement the locality of currently used trigram models. The structured language model, its probab...

In this article we present the realization of a generic finite-state system. The system has been used to create concrete lexical tools for word form analysis, word form generation, creation and derivation history, and spenchecking. It will also be used to create a finite-state transducer for the recognition of phrases. Producing a finitestate component with the generic system requires little e~...