E xternal financing is critical to ventures that do not have a revenue source but need to recruit employees, develop products, pay suppliers, and market their products/services. There is an increasing belief among entrepreneurs that electronic word-of-mouth (eWOM), specifically blog coverage, can aid in achieving venture capital financing. Conflicting findings reported by past studies examining eWOM make it unclear what to make of such beliefs of entrepreneurs. Even if there were generally agreed-upon results, a stream of literature indicates that because of the differences in traits between the prior investigated contexts and venture capital financing, the findings from the prior studies cannot be generalized to venture capital financing. Extant studies also fall short in examining the role of time and the status of entities generating eWOM in determining the influence of eWOM on decision making. To address this dearth of literature in a context that attracts billions of dollars every year, we investigate the effect of eWOM on venture capital financing. This study entails the challenging task of gathering data from hundreds of ventures along with other sources including VentureXpert, surveys, Google Blogsearch, Lexis-Nexis, and Archive.org.The key findings of our econometric analysis are that the impact of negative eWOM is greater than is the impact of positive eWOM and that the effect of eWOM on financing decreases with the progress through the financing stages. We also find that the eWOM of popular bloggers helps ventures in getting higher funding amounts and valuations. The empirical model used in this work accounts for inherent selection biases of entrepreneurs and venture capitalists, and we conduct numerous robustness checks for potential issues of endogeneity, selection bias, nonlinearities, and popularity cutoff for blogs. The findings have important implications for entrepreneurs and suggest ways by which entrepreneurs can take advantage of eWOM.

Surface anatomy is an integral part of medical education and enables medical students to learn skills for future medical practice. In the past decade, there has been a decline in the teaching of anatomy in the medical curriculum, and this study seeks to assess the attitudes of medical students to participation in surface anatomy classes. Consequently, all first year medical students at the Guy's, King's and St Thomas's Medical School, London, were asked to fill in an anonymous questionnaire at the end of their last surface anatomy session of the year. A total of 290 medical students completed the questionnaires, resulting in an 81.6% response rate. The students had a mean age of 19.6 years (range 18-32) and 104 (35.9%) of them were male. Seventy-six students (26.2%) were subjects in surface anatomy tutorials (60.5% male). Students generally volunteered because no one else did. Of the volunteers, 38.2% would rather not have been subjects, because of embarrassment, inability to make notes, or to see clearly the material being taught. Female medical students from ethnic minority groups were especially reluctant to volunteer to be subjects. Single-sex classes improved the volunteer rate to some extent, but not dramatically. Students appreciate the importance of surface anatomy to cadaveric study and to future clinical practice. Computer models, lectures, and videos are complementary but cannot be a substitute for peer group models, artists' models being the only alternative.

Abstract-The variable nature of the wireless channel may cause the quality of service to be intolerable for certain applications. To combat channel variability, we consider rate adaptation at the physical layer. We build an adaptive communication system based on uncoded QAM in which the available information on the channel state is obtained using the mere packet-level ACK/NACK sequence. Our system chooses the constellation size that maximizes the expected packet level goodput for every single packet. Our simulations show that our system achieves a goodput, reasonably close to the highest possible goodput achievable with full-feedback on Rayleigh-fading Markov channels.

In this work, we followed the sentiment analysis literature, and used supervised learning methods, which take manually classified data (corpus) as input and automatically extract features (combination of words and parts of speech of words) for sentiment analysis (Dave et al. 2003; Ghose and Ipeirotis 2011; Pang et al. 2002; Shanahan et al. 2006). These supervised methods do not rely on manually or semi-manually constructed discriminant-word lexicons. Prior research has shown that supervised methods perform better than lexicon-based approaches for sentiment analysis (Chaovalit and Zhou 2005; Pang et al. 2002).

Abstract-We consider the problem of simultaneous user-scheduling, power-allocation, and rate-selection in an orthogonal frequency division multiple access (OFDMA) downlink, with the goal of maximizing expected sum-utility under a sum-power constraint. In doing so, we consider a family of generic goodput-based utilities that facilitate, e.g., throughput-based pricing, quality-of-service enforcement, and/or the treatment of practical modulationand-coding schemes (MCS). Since perfect knowledge of channel state information (CSI) may be difficult to maintain at the base-station, especially when the number of users and/or subchannels is large, we consider scheduling and resource allocation under imperfect CSI, where the channel state is described by a generic probability distribution. First, we consider the "continuous" case where multiple users and/or code rates can time-share a single OFDMA subchannel and time slot. This yields a nonconvex optimization problem that we convert into a convex optimization problem and solve exactly using a dual optimization approach. Second, we consider the "discrete" case where only a single user and code rate is allowed per OFDMA subchannel per time slot. For the mixed-integer optimization problem that arises, we discuss the connections it has with the continuous case and show that it can solved exactly in some situations. For the other situations, we present a bound on the optimality gap. For both cases, we provide algorithmic implementations of the obtained solution. Finally, we study, numerically, the performance of the proposed algorithms under various degrees of CSI uncertainty, utilities, and OFDMA system configurations. In addition, we demonstrate advantages relative to existing state-of-the-art algorithms.Index Terms-Bisection algorithm, imperfect channel state information (CSI), mixed integer optimization, orthogonal frequency division multiple access (OFDMA), resource allocation, scheduling, utility maximization.