When we are looking for information about a specific person, we type the name in search engines. Then, search engines return many web pages which include the given name strings. However, the resulting web pages are mixed with related information and unrelated information. This is called a name disambiguation problem. Current search engines provide links but do not distinguish related and unrelated web pages. In this paper, we described our algorithm to solve a name disambiguation problem using two linear algebraic approaches: Singular Valued Decomposition (SVD) and Nonnegative Matrix Factorization (NMF). Experiments show that using NMF algorithm yields the best performance in terms of precision, recall and Fmeasure, and applying SVD requires only 10% computation time compared to traditional K-means algorithm. Our solution with search engines will provide more precise search results than traditional search engine services.