• London School of Economics and Political Science Featured PhD Programmes
  • University of Leeds Featured PhD Programmes
  • University of Leeds Featured PhD Programmes
  • University of Leeds Featured PhD Programmes
  • Carlos III Health Institute Featured PhD Programmes
  • University of Glasgow Featured PhD Programmes
  • University of Mannheim Featured PhD Programmes
University of York Featured PhD Programmes
University of Westminster Featured PhD Programmes
University of Bristol Featured PhD Programmes
University of Glasgow Featured PhD Programmes
University of Reading Featured PhD Programmes

Exploring Sparse Random Projections for Streaming Algorithms

This project is no longer listed in the FindAPhD
database and may not be available.

Click here to search the FindAPhD database
for PhD studentship opportunities
  • Full or part time
    Dr Cadenas
    Dr Stahl
  • Application Deadline
    Applications accepted all year round
  • Self-Funded PhD Students Only
    Self-Funded PhD Students Only

Project Description

Project Overview:

Reducing the number of dimensions in data, at a high level, is all about stripping out unimportant parts of data to express and amplify the important ones. The procedure aims at turning the data in such a way as to make it easier to process and to improve the quality of knowledge that is learned from it by any further processing on that data. There are many dimensionality reduction algorithms and techniques.
Data itself mostly dictates what techniques to use. For example, for text documents, Singular Value Decomposition is used extensively. When the time to process the data is critical, such as in on-line streaming algorithms, performance of any dimensionality reduction algorithm is paramount. Unfortunately, most common and best known dimensionality techniques such as PCA, Approximate PCA, etc. are not fast.
This project aims to investigate a relatively unknown technique referred to as Sparse Random Projection. It is claimed the technique is fast, easy to implement and easy to understand. But what does Sparse Random Projection have to offer, for streaming algorithms? The project will quantitatively evaluate, using common streaming algorithm as benchmarks, the Sparse Random Projection technique in particular.
From there search for potential improvements towards providing contributions for fast dimensionality reduction algorithms suitable for streaming computation will be sought.

School of Systems Engineering, University of Reading:
The University of Reading is one of the UK’s 20 most research-intensive universities and among the top 200 universities in the world. Achievements include the Queen’s Award for Export Achievement (1989) and the Queen’s Anniversary Prize for Higher Education (1998, 2006 and 2009). This project will take place in the School of Systems Engineering (SSE), which has a strong reputation for its innovative research in computer science, cybernetics, and electronic engineering.

How to apply:
(1) Submit an application for a PhD in Computer Science using the link below.
(2) After submitting your application you will receive an email to confirm receipt; email should be forwarded along with a covering letter and full CV to Dr Oswaldo Cadenas ([email protected]).
(3) In the online application system, there is a section for “Research proposal” and a box that says “If you have already been in contact with a potential supervisor, please tell us who” – in this box, please enter “Dr. Oswaldo Cadenas”.

Funding Notes

We welcome applications from self-funded students worldwide for this project.
Students from Brazil: we welcome and support applications for the Science Without Borders Scholarship (Ciência sem Fronteiras).

Applicants should have a bachelors (at least 2.1 or equivalent) or masters degree in Computer Science or a strongly related discipline. Strong analytical and logic skills are preferable. Experience in programming is desirable, C, C++, Python or CUDA.

Share this page:

Cookie Policy    X