Withdraw
Loading…
Mining LEGO data sets to support LEGO design
Zhang, Xiaodan
Loading…
Permalink
https://hdl.handle.net/2142/89051
Description
- Title
- Mining LEGO data sets to support LEGO design
- Author(s)
- Zhang, Xiaodan
- Issue Date
- 2015-12-07
- Director of Research (if dissertation) or Advisor (if thesis)
- Zhai, ChengXiang
- Department of Study
- Computer Science
- Discipline
- Computer Science
- Degree Granting Institution
- University of Illinois at Urbana-Champaign
- Degree Name
- M.S.
- Degree Level
- Thesis
- Keyword(s)
- Data Mining
- Information Retrieval
- Knowledge Discovery
- LEGO
- Abstract
- "The Lego Group invented LEGO bricks. These bricks could be put up together to build creative LEGO sets of different themes, such as a Volkswagen T1 Camper Van model (in ``Sculpture"" theme) or The Simpsons House model (in ``Town"" theme). LEGO accompanies nearly everyone from youth to adulthood. The age groups of fans range from kids in pre-school to elder people who have grandchildren. With such a strong and huge fan base, there appear a lot of websites that provide LEGO data of sets, parts, minifigures as well as online communities for fans to share their experience about LEGO sets. However, there barely exists any research in this rich data source to discover knowledge and insights about how each LEGO part plays a role in a LEGO set, in its own part category and in a LEGO theme; how each LEGO set is different from the other one in the aspects of theme and part components. There are a lot of interesting questions we can address from the datasets that will not only help better LEGO designs, but will also help LEGO fans or potential customers make efficient purchasing decisions when they get more familiar with LEGO sets and parts. To address these needs, in this thesis, we propose a systematic method of mining LEGO datasets of sets and parts to support LEGO design. Treating each LEGO set as a document and each part in it as a word, we are able to apply data mining techniques, such as Topic Model and K-Means Clustering to find statistics of sets and parts. The preliminary experiment results show that the proposed methods can automatically construct a LEGO Brick Lexicon that shows a part's relationship with other parts, sets and themes, discover knowledge about typical LEGO construction patterns and create hybrid theme recommendations. We believe this is a step forward to helping LEGO designers create more attractive sets with pragmatic parts as well as improving the building experience of LEGO fans/builders."
- Graduation Semester
- 2015-12
- Type of Resource
- text
- Permalink
- http://hdl.handle.net/2142/89051
- Copyright and License Information
- Copyright 2015 Xiaodan Zhang
Owning Collections
Graduate Dissertations and Theses at Illinois PRIMARY
Graduate Theses and Dissertations at IllinoisDissertations and Theses - Computer Science
Dissertations and Theses from the Dept. of Computer ScienceManage Files
Loading…
Edit Collection Membership
Loading…
Edit Metadata
Loading…
Edit Properties
Loading…
Embargoes
Loading…