![](https://s32615.pcdn.co/wp-content/uploads/2020/08/Zeiberg_Daniel_Hero-scaled.jpg)
![Need to add an ALT text field to the CMS for the Hero Image](https://s32615.pcdn.co/wp-content/uploads/2020/08/Zeiberg_Daniel_Hero-mobile.jpg)
Daniel Zeiberg
![](https://s32615.pcdn.co/wp-content/uploads/2020/08/Zeiberg_Daniel_Hero-scaled.jpg)
![Daniel Zeiberg](https://s32615.pcdn.co/wp-content/uploads/2020/08/Zeiberg_Daniel_Hero-mobile.jpg)
Daniel Zeiberg is a PhD student at Northeastern University’s Khoury College of Computer Sciences, advised by Predrag Radivojac. He received his BSE in Computer Science from the University of Michigan, where he did undergraduate research developing machine learning methods with applications in clinical medicine.
His research focuses on the development of machine learning methods and their applications in bioinformatics. He focuses on learning in the positive-unlabeled classification setting, a semi-supervised classification setting in which one is interested in learning a binary classifier from a small set of labeled positives and a large set of unlabeled data. This problem setting arises in a variety of applications, including bioinformatics. For example, in one of his current projects, predicting the pathogenicity of genetic mutations, datasets often include a large set of known pathogenic mutations but relatively few known benign mutations. Through methods that the field has developed, it is possible to use learn a classifier that can predict whether the unlabeled mutations cause a disease without seeing any labeled benign mutations.
What Daniel finds most interesting about his research is that the applications he works on will one day have a positive impact on patients. Moving forward, he would like to develop tools that are used to improve patient outcomes.
I am advised by Predrag Radivojac. In my current project I am developing machine learning methods to estimate the class prior in positive-unlabeled data.
I am interested in developing machine learning methods and applying them to bioinformatics problems.
I would like to research methods that can handle irregularly sampled time series.
What I, and I think others should, find most interesting about my research is that I develop applications that have a positive impact on people.
My career goal is to implement novel machine learning techniques in high impact products and applications.
South Jersey.
I received my undergraduate degree from the University of Michigan, where I studied computer science and statistics. I liked Michigan’s school spirit, as I came from a small high school.
Daniel Zeiberg is a PhD student at Northeastern University’s Khoury College of Computer Sciences, advised by Predrag Radivojac. He received his BSE in Computer Science from the University of Michigan, where he did undergraduate research developing machine learning methods with applications in clinical medicine.
His research focuses on the development of machine learning methods and their applications in bioinformatics. He focuses on learning in the positive-unlabeled classification setting, a semi-supervised classification setting in which one is interested in learning a binary classifier from a small set of labeled positives and a large set of unlabeled data. This problem setting arises in a variety of applications, including bioinformatics. For example, in one of his current projects, predicting the pathogenicity of genetic mutations, datasets often include a large set of known pathogenic mutations but relatively few known benign mutations. Through methods that the field has developed, it is possible to use learn a classifier that can predict whether the unlabeled mutations cause a disease without seeing any labeled benign mutations.
What Daniel finds most interesting about his research is that the applications he works on will one day have a positive impact on patients. Moving forward, he would like to develop tools that are used to improve patient outcomes.
I am advised by Predrag Radivojac. In my current project I am developing machine learning methods to estimate the class prior in positive-unlabeled data.
I am interested in developing machine learning methods and applying them to bioinformatics problems.
I would like to research methods that can handle irregularly sampled time series.
What I, and I think others should, find most interesting about my research is that I develop applications that have a positive impact on people.
My career goal is to implement novel machine learning techniques in high impact products and applications.
South Jersey.
I received my undergraduate degree from the University of Michigan, where I studied computer science and statistics. I liked Michigan’s school spirit, as I came from a small high school.