Se June Hong

Title: Innovations of Se June Hong

Introduction

Se June Hong is a notable inventor based in Yorktown Heights, NY (US). He has made significant contributions to the field of predictive modeling, holding a total of 13 patents. His work focuses on enhancing the accuracy and interpretability of predictive models, which has important implications in various applications, including data mining and marketing.

Latest Patents

One of his latest patents is titled "Method for ensemble predictive modeling by multiplicative adjustment of class probability: APM." This innovative method models class probability from data using a novel multiplicative adjustment based on multiple items of evidence derived from training data. The optimal adjustment factors can be determined through various techniques, with maximum likelihood being a preferred method. This approach allows for effective aggregation of probabilistic evidence, resulting in a model output that exhibits a low error rate for classification. It also provides a superior lift curve for distinguishing between classes, making it valuable for data mining and marketing applications.

Another significant patent is "Cascade boosting of predictive models." This method addresses the interpretability issues of previous boosting techniques while reducing fragmentation when applied to decision trees. Cascade boosting organizes weak models in a decision list rather than a weighted average, simplifying the predictive model and enhancing the understanding of data correlations. This method can be applied to both segmented and non-segmented predictive models, resulting in fewer rules and tree leaves.

Career Highlights

Se June Hong is currently employed at International Business Machines Corporation (IBM), where he continues to innovate in the field of predictive modeling. His work has garnered attention for its practical applications and contributions to the advancement of data analysis techniques.

Collaborations

Some of his notable coworkers include Ravindra K Nair and Daniel Lawrence Ostapko, who have collaborated with him on various projects within IBM.

Conclusion

Se June Hong's contributions to predictive modeling through his innovative patents have significantly advanced the field. His work not only enhances the accuracy of predictions but also improves the interpretability of complex models.