Blessed by the Internet age, many online retailers (e.g., Amazon.com) have deployed recommender systems to help their customers identify products that may be of their interest in order to improve cross-selling and enhance customer loyalty. Collaborative Filtering (CF) is the most successful technique among different approaches to generating recommendations. Collaborative filtering automates the word-of-mouth process of recommendations by aggregating item ratings of other like-minded users. Although it has achieved huge success in both research and industry, traditional CF suffers a fundamental problem because of its dependence on users' numeric ratings as its sole source of user preference information. However, user ratings are often unable to fully represent user preferences. As a result, it is difficult to identify the true similarity among users based on ratings only. On the other hand, people are more and more comfortable with expressing themselves online using free-form text (e.g., blogs). As a result, user text reviews on different products have become a common source for consumers to share and receive information about products. Recent studies have also found that user reviews have significant impact on online businesses' sales. Inspired by the results of these studies, I propose an integrated approach to collaborative filtering called urCF (User Review enhanced Collaborative Filtering) that integrates user text reviews into memory-based collaborative filtering in order to better model users' preferences and in turn enhance the performance of CF-based recommender systems. This research uses existing text mining techniques to extract user opinions on item features. It proposes a new weighting scheme based on TF-IDF to measure the priority of item features in influencing users' overall opinions of different items. This study also explores and compares several different approaches to integrating user opinion information extracted from user text reviews into the user similarity measurement of traditional CF-based recommender systems. The proposed urCF system is evaluated against existing approaches using a dataset collected from Yahoo! Movies. The results show that the proposed urCF system significantly improved the performance of memory-based recommender systems. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]