Physical Sciences and Mathematics Commons^™

Mining Concept In Big Data, Jingjing Yang

Master's Projects

To fruitful using big data, data mining is necessary. There are two well-known methods, one is based on apriori principle, and the other one is based on FP-tree. In this project we explore a new approach that is based on simplicial complex, which is a combinatorial form of polyhedron used in algebraic topology. Our approach, similar to FP-tree, is top down, at the same time, it is based on apriori principle in geometric form, called closed condition in simplicial complex. Our method is almost 300 times faster than FP-growth on a real world database using a SJSU laptop. The database …