Detection Method to Prevent Reference Omissions Using Logs of Text Inserted by Copy-and-Paste Operations
Chikako Ishizawa, Masahiro TOSHIMA, Yoichi Kageyama, Makoto Nishida · International Journal of the Society of Materials Engineering for Resources · 2018
The objective of this study is to prevent human mistakes, specifically, omitting reference lists and reference numbers that should be appended to a copied text. When sentences are created using a personal computer (PC), it is possible to copy and use text from the Internet. However, a user will occasionally forget to include the copy-source information, e.g., the URL or webpage title, in the reference list or to add the reference number to the inserted text. Therefore, we propose a method that obtains the relevant logs from the PC, detects the text inserted by a copy-and-paste operation, and shows the copy-source information to the user. The proposed method consists of three processes. The first process monitors the clipboard state and creates logs with the copy-source information, e.g., the webpage title or the URL. Thereafter, the text-detection process calculates the cosine similarity between the copy-source sentences and the user's text. Text with no reference information is detected based on the cosine-similarity value. Finally, the reference information for the detected text is displayed on the monitor. Various changes were made to the inserted text by using Microsoft Windows 7. Our experimental results showed that the proposed method can detect inserted text without reference information and display the necessary information in the reference list.