著者

下窪 聖人

出版者

法政大学大学院情報科学研究科

雑誌

法政大学大学院紀要. 情報科学研究科編 (ISSN:24321192)

巻号頁・発行日

vol.16, pp.1-6, 2021-03-24

---

In recent years, E-book provides many opportunities that people read books during a commute or break time. Though an E-book reader provides some useful functions for reading the books. This system enables a new function of recommending illustrations during one is reading. This is achieved by calculating cosine similarity which uses the BERT based distributed representation of each scene. There are mainly 4 steps: (1) this system divides texts into some scenes by TopicTiling; (2) the important sentences are extracted by using BERT for each scene; (3) subjects, verbs, and objects are extracted using MeCab and CaboCha; and (4) the system recommends appropriate illustrations by calculating similarity between the sentences by using the analysis results and image titles. As expected, this system can recommend illustrations for a reader to form the image when he is difficult to understand E-book in only text content. These recommended illustrations are original from the illustrations of the web site “irasutoya”. As a result of the experiment, the accuracy of the illustration’s recommendation was about 72.0 percent. This result was obtained by using CNN to convert the image style of both the original and the recommended illustrations, and then evaluating them by SSIM.

著者

松田 達希

出版者

法政大学大学院情報科学研究科

雑誌

法政大学大学院紀要. 情報科学研究科編 = 法政大学大学院紀要. 情報科学研究科編 (ISSN:24321192)

巻号頁・発行日

no.11, pp.1-6, 2016-03-24

---

This paper describes a query language and a complex event processing (CEP) system over multiple continuous XML data streams. First, multiple stream processing is formulated in algebraic expressions for stream filtering, union, activation, decomposition and partition. Next, a query language, called QLMXS, over XML streams is defined on all functions of the algebraic models in an SQL-like form. Finally, a CEP system is proposed for processing QLMXS queries. The performance of the system is dominated by parsing XML documents and finding specific sequences of events. The former is resolved by adopting VTD parser, which is a modern XML parser and faster than the traditional technologies of DOM, SAX and StAX. The latter is achieved by a finite state automaton (FSA) that determines the acceptance of the event sequences represented in QLMXS queries. Experimental results demonstrate that the performance of the proposed QLMXS system becomes 1.7 times faster than the previous implementation which used the StAX parser and the visibly pushdown automaton (VPA) as cores of the analytical engines.