Building open-domain conversational systems (or chatbots) that produce convincing responses is a recognized challenge. Recent state-of-the-art (SoTA) transformer-based models for the generation of natural language dialogue have demonstrated impressive performance in simulating human-like, single-turn conversations in English. This work investigates, by an empirical study, the potential for transfer learning of such models to Swedish language. DialoGPT, an English language pre-trained model, is adapted by training on three different Swedish language conversational datasets obtained from publicly available sources. Perplexity score (an automated intrinsic language model metric) and surveys by human evaluation were used to assess the performances of the fine-tuned models, with results that indicate that the capacity for transfer learning can be exploited with considerable success. Human evaluators asked to score the simulated dialogue judged over 57% of the chatbot responses to be human-like for the model trained on the largest (Swedish) dataset. We provide the demos and model checkpoints of our English and Swedish chatbots on the HuggingFace platform for public use.
<span>The recent advances in technology, the increased dependence on electrical energy and the emergence of the fourth industrial revolution (Industry 4.0) were all factors in the increased need for smart, efficient and reliable energy systems. This introduced the concept of the Smart Grid (SG). A SG is a potential replacement for older power grids, capable of adapting and distributing energy based on demand. SG systems are complex. They combine various components and have high requirements for real time reliable operation. This paper attempts to provide an overview of SG systems, by outlining SG architecture and various components. It also introduces communication technologies, integration and network management tools that are involved in SG systems. In addition, the paper highlights challenges and issues that need to be addressed for a successful implementation of SG. Finally, we provide suggestions for future <br /> research directions. </span>
As the Bitcoin keeps increasing in value compared to other cryptocurrencies, more attention has given to Blockchain Technology (BT), which is the infrastructure behind the Bitcoin, especially on its role in addressing the problems of the classical centralized system. As a digital currency, Bitcoin is dependent on the decentralized cryptographic tools and peer-to-peer system. The digital currency implements a distributed ledger using Blockchain when verifying any type of transaction. In this paper, the aim is to describe how digital currency networks such as Bitcoin provides a "trust-less" platform for users to embark on money transfers without necessarily depending on any central trusted establishments such as payment services or financial institutions. Furthermore, this work comprehensively overviewed the basic principle that underly BT, such as transaction, consensus algorithms, and hashing. This study also provided a novel classification for blockchain types according to their system architecture and consensus strategy. For each type, our contribution was provided with an example, which clearly describes the blockchain features and the transaction steps. Our classification intended to help researchers understand and choose the blockchain for their application. The paper ends with the discussion of the differences between each type
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.