부산대학교 도서관

When we attempt to describe an ecommerce data, we usually visualize it in the form of name value pairs. Every item is identified and defined by a set of specifications that it gets associated with. When a key gets its specifications, it gets attributes to get compared with. A recommendation system can always look for similar specifications before making one. The data of a person or a place is limitedly seen as a name value pair. Sites like Wikipedia address a few characteristics of a personality in associative pairs and others mostly as textual description. Through this research, we propose a model to build a name-value pair for the people and place data. We design a model which first captures all the basic data of different eminent personalities and observes through the common pool of specifications. After having a common set of specifications, data is crawled and parsed from Wikipedia to complete the missing entries. Thus forms a PD-Box, box data visualization for people and place data. This data set representation can be used in processing engines to compare and evaluate recommendations to combine people and place with every other entity data. The results presented appear to be promising to combine this model with processing and search engines. The model uncovers and unwraps hidden meaningful data about people and place which can be an insightful direction for research and data structuring.