We describe the 2017 version of Microsoft's conversational speech recognition system, in which we update our 2016 system with recent developments neural-network-based acoustic and language modeling to further advance state art on Switchboard task. The adds a CNN-BLSTM model set architectures combined previously, includes character-based dialog session aware LSTM models rescoring. For combination adopt two-stage approach, whereby subsets are first at senone/frame level, followed by word-level...

Great changes within the business environment in last 2 decades have made supply chain management (SCM) an important topic for academics and practitioners. Globalization, larger reliance on layered suppliers specialized capabilities innovation, changing customer needs, networks to maintain a competitive advantage, developments technology all contributed very different environment. Most current works view customers as being outside design; however, we believe that are missing link chain....

Purpose The purpose of this paper is to review research on options theory (OT) in the management information systems (MIS) field. Design/methodology/approach reviews OT and suggests further ways it may be useful for MIS. Findings explores current global trends past uses by MIS researchers within a framework. objective broaden discussion field, facilitate development, close gaps existing knowledge, uncover areas where needed. It believed that will benefit from study. Research...

Abstract This paper presents a two‐phase quantitative approach for enhanced index investing based on the mean‐variance model and goal programming method. In first stage, we use theory to select better performing stocks an investment pool. Then, in second method weight selected by balancing both tracking error rate of return. addition theoretical formulation, construct spreadsheet‐based decision support system (DSS) transaction data help resolve problem. The contributes literature two ways....

The paper contributes to the literature in two ways. For academics, we present original discussions on combining an interdisciplinary mean variance model and a goal programming method. practitioners, discussion using DSS support index investing. results of empirical survey Taiwan stock market are also presented.

The evaluation of the high tech Thin Film Transistor-Liquid Crystal Display (TFT-LCD) industry is not easy. Decisions about investing in latest generation plant involve billions dollars and a great deal uncertainty. traditional company method Net Present Value (NPV) rule 'static discount future cash flow' concept that fails to capture characteristic uncertain enormous jump revenue when price war arises, which are critical this industry. We incorporate unique characteristics TFT-LCD into our...

Service Science, Management, and Engineering (SSME) is an emerging interdisciplinary approach for the learning of customer needs. In this paper, we review related literature theories, explore tools used in various fields relevant. Then, propose a systematic three-phase framework SSME research.

Most current works in supply chain view customers as being outside the design; however, this paper, we focus on missing link of customer's role chain. We present an original discussion emerging discipline services science, management, and engineering (SSME) SCM, identify bridge gaps between two disciplines, discuss ways to align them. In addition, consider specific challenges well opportunities for SSME field.

In this study, we used a combined stochastic process and value-at-risk (VaR) method to examine an electronic commerce expansion decision. By modeling uncertain benefits as process, maximum losses of alternative decisions were quantified compared help managers make information system/information technology (IS/IT) project decisions. Our results, based on the loss perspective, demonstrated that uncertainty plays critical role in evaluating IS/IT projects. More importantly, results illustrate...