Search

Bibliographic references containing citation information of academic literature play an important role as a medium connecting earlier and recent studies. As references contain machine-readable metadata such as author name, title, or publication year, they have been widely used in the field of citation information services including search services for scholarly information and research trend analysis. Many institutions around the world manually extract and continuously accumulate reference metadata to provide various scholarly services. However, manually collection of reference metadata every year continues to be a burden because of the associated cost and time consumption. With the accumulation of a large volume of academic literature, several tools, including GROBID and CERMINE, that automatically extract reference metadata have been released. However, these tools have some limitations. For example, they are only applicable to references written in English, the types of extractable metadata are limited for each tool, and the performance of the tools is insufficient to replace the manual extraction of reference metadata. Therefore, in this study, we focused on constructing a high-quality corpus to automatically extract metadata from multilingual journal article references. Using our constructed corpus, we trained and evaluated a BERT-based transfer-learning model. Furthermore, we compared the performance of the BERT-based model with that of the existing model, GROBID. Currently, our corpus contains 3,815,987 multilingual references, mainly in English and Korean, with labels for 13 different metadata types. According to our experiment, the BERT-based model trained using our corpus showed excellent performance in extracting metadata not only from journal references written in English but also in other languages, particularly Korean. This corpus is available at http://doi.org/10.23057/47.

Abstractive summarization is challenging problem, especially abstractive summarization based on unsupervised learning because it must generate whole, unique sentences. In the real world, companies use abstractive summarization to understand customer feedbacks. In many cases, this work is done by humans and so is expensive in terms of time and money. Therefore, there is an increasing demand for machine learning-based abstractive summarization systems. However, most previous abstractive summarization studies were of supervised models. In this paper, we proposed novel abstractive summarization methods that can be trained unsupervisedly. One of the proposed methods is based on Adversarially Regularized Autoencoder(ARAE) model, but abstractive summary generation method for each cluster of similar customers’ reviews, is newly proposed. We further proposed Conditional Adversarially Regularized Autoencoder(CARAE) model which is similar to the ARAE model but with the addition of condition nodes so that additional information about the cluster can be used during summarization. We first performed summary experiments based on Korean and additionally performed experiments on English. In the experiments, we set up some comparison models and used ROUGE and BLEU to evaluate our proposed models’ performance. Overall, our proposed models outperformed the comparison models and CARAE model performed better than the ARAE model.

In general, cloud environments are based on the pay-per-use model, whereby clients pay for the information resources provided by cloud service providers. Users rent and use resources as needed while avoiding the high costs of large-scale resource acquisitions, and providers maximize their profits by managing information resources at a minimum cost while upholding service-level agreements. As the number of resources gradually increases, power supply shortages may arise. This study focuses on the fact that the CPU utilization rate of the server running in the data center is less than 30% and idle servers running only the OS consume more than half of the power consumed by hosts running with maximum CPU utilization and speed. Therefore, this study proposes an approach to enhance data center efficiency through improved management of energy consumption. We present a method to minimize energy consumption while processing the same workload, i.e., ultimately reducing the energy consumed by operating servers. Energy consumption and SLA violation rate were used as evaluation metrics of the optimization model to meet the minimum performance target.

It is often very difficult to locate information on the Web because of its large and rapidly increasing amount of data. One key reason for this is traditional keyword-based search engines focus only on the resources whose title or content exactly matches the query keywords. People usually want to find the best matching resource itself to their query, not the documents which contain the resource. Recently, one promising way to meet this kind of requirement must be ontology-based approach for semantic search. However, it is also obvious there is still non-negligible gap between average users and ontological approach. To overcome this limitation of ontological approach such as Semantic Web, it is essential to provide an efficient method to fill the gap while taking full advantage of semantic technologies. To this end, we devise a method to generate alternative SPARQL queries from the typical natural language based query to the conventional search engines and evaluate the most matched SPARQL query among the alternatives by considering the characteristics of the target knowledge bases. We then implement a prototype system to evaluate the proposed method and validate its empirical performance and accuracy.