Skip to main content


Office Location

440 Huntington Avenue
266 West Village H
Boston, MA 02115


Saber Shokat Fadaee is a PhD student in the Computer Science program at Northeastern University’s College of Computer and Information Science, advised by Professor Ravi Sundaram. Before coming to Northeastern, Saber earned his Master of Science degree in Software Engineering from Sharif University of Technology and his Bachelor of Science degree in Computer Science from the University of Tehran. Saber’s research is focused on machine learning and data mining, and he is passionate about working with real world data. At Northeastern, he works on projects which help to solve certain classification problems, and he participates in the Applied Machine Learning Group.


  • MS in Software Engineering, Sharif University of Technology – Iran
  • BS in Computer Science, University of Tehran – Iran

About Me

  • Hometown: Boston, Massachusetts
  • Field of Study: Computer Science
  • PhD Advisor: Ravi Sundaram

What are the specifics of your graduate education (thus far)?

I started my PhD in the theory group, but soon changed my research to machine learning and data mining. Since then, I’ve been passionate about working with real world data. One of the most interesting problems in machine learning that I’ve been dealing with is the problem of “classification”. The goal in classification is to learn a mapping from input “X” to output “Y”. Classification becomes more challenging when the algorithm is not guaranteed that every measurement in its input vector is always provided. In other words, when the machine learning algorithm is given a new example “X” but some entries of “Xi” is missing, we should find a way to impute the missing values. Our algorithm must provide a prediction of the values of the missing entries.

What are your research interests?

In my first data scientific project, I dealt with an interesting classification problem in which we tried to classify graphs that represent “persons of interest” into five different categories using their network structures.

In my second project, I’m going to solve a classification problem with missing values with application in recommendation systems. In recommendation systems to goal is to predict the “rating” or “preference” that a user would give to an item.

In my third project, we are given an interesting data set of bot-nets, and we are going to propose a few methods to convert this problem into a familiar problem that we studied before: classification with missing values.

What’s one problem you’d like to solve with your research/work?

I’ve always been fascinated by this question: what are the fundamental statistical or computational or theoretic laws that govern all learning systems, including computers and humans? More specifically, I’d like to know how we make a decision when we want to purchase an item, and apply that information to improve the performance of the state-of-the-art recommendation systems.

What aspect of what you do is most interesting?

I’ve been looking for features that can best represent the data. Whether it’s a network of people, or the list of ratings some users have given to a list of movies, or information about bot-net behaviors. I’ve discovered for example, that in a network of people the small groups of highly connected people, can say a lot about that network and can reconstruct the whole network together.

What are your research or career goals, going forward?

I’d like to apply my knowledge on the real world challenges in the industrial section, for a little while, and then return to academia with the strong background and experiences that I accumulated in that sector.