The intent recognition is an essential algorithm of any conversational AI application. It responsible for the classification input message into meaningful classes. In many bot development platforms, we can configure NLU pipeline. Several services are currently available as API, or choose from open-source alternatives. However, there no comparison and algorithms. Many factors make selection righ...

Proceeding from the scattered field expansion in cylindrical harmonics, we define partial invisibility or cloaking as suppression of finitely many lowest-order harmonics. We show that for benchmark structures (a dielectric rod and a perfectly conducting cylinder circular cross section covered by concentric layer), such multiple can be achieved. For this purpose, prove solvability explicitly det...

Data envelopment analysis (DEA) with considering the best condition for each decision making unit (DMU) assesses the relative efficiency for it and divides a homogenous group of DMUs in to two categories: efficient and inefficient, but traditional DEA models can not rank efficient DMUs. Although some models were introduced for ranking efficient DMUs, Franklin Lio & Hsuan peng (2008), proposed a...

  This paper proposes a suitable benchmark for inefficient commercial bank branches by using Data Envelopment Analysis (DEA). In order to render an inefficient bank branch efficient, it is necessary to decrease inputs and increase outputs. As there are priorities for decreasing some certain inputs and increasing some certain outputs over other inputs and outputs, respectively, it is necessary ...

Although DEA is a powerful method in evaluating DMUs, it does have some limitations. One of the limitations of this method is the result of the evaluation is based on previously data and the results are not proper for forecasting the future changes. So For this purpose, we design feedback loops for forecasting inputs and outputs through system dynamics and simulation. Then we use DEA model to f...