The increasing amount of text found in all areas including articles, magazines, research papers, academic seminars, etc. calls for a proper way to easily and effectively interpret large volumes of text. From these requirements, word cloud has emerged as one of the new techniques for obtaining and displaying the most meaningful words.
Word cloud is a form of graphical display that allows users to easily communicate with text, documents, etc., so that they can easily and efficiently understand them in a short time. In the word cloud, high-weight words are placed preferentially. That is, the heaviest word is put in the center, around which the remaining ones are placed. To visualize word clouds, many visual attributes such as font size and color were used. Here, font size was given prior importance. Frequent terms are easily found in the word cloud due to their large font size.
Generally, many tabular forms are used to visualize the content of a database. This form is effective for reflecting a certain situation, but it does not accurately reflect the degree of association between data and, moreover, it is difficult to visualize the degree of change in the data as it is shown continuously.
Jong Chol Sam, a researcher at the Faculty of Information Science and Technology, has proposed an approach to optimize cloud placement of words based on the estimation of similarity between words using the structure of database and the clusterizing by similarity matrix.
First, he performed a similarity evaluation using database structure and carried out a clustering by similarity matrix of words. Then, he determined the font size and bounding box from the weight of words and optimized the word layout on the 2D canvas.
The experimental results show that when the content of database is displayed as a word cloud by the proposed method, users can quickly acquire information that reflects the semantic relations of data and a lot of information can be displayed in less space.