Sherlock is an NLeSC-wide project in which all eScience Research engineers of the NLeSC work together. The goals of Sherlock are:
- Team building: get to know your (new) colleagues, their interests, skills, knowledge, etc.
- Knowledge sharing & development: learn from each other and collectively learn new skills.
- eStep: producing generic tools, knowledge and research for eStep.
- Boost the parent project: give a project access to the skills and knowledge 30 engineers.
For Sherlock the parent project is A Jungle Computing Approach to Large Scale Online Forensic Analyis, a cooperation with the NFI and the HPDC group at VU University.
Sherlock is about Digital Forensics, more specifically about finding digital evidence in large amounts of data. Finding such evidence requires analyis of large amounts of text and multimedia data and means of clustering, visualising and searching through the results.
The Sherlock teams will work on the following topics:
- Analyzing large document collections
- Concept search
- Deep learning for computer vision
- Visualization of timelines
- Cluster analysis
- Tool and data integration
All sprints will consist of three consecutive days (Tuesday to Thurday). The dates are:
- 20-22 October (agenda)
- 1-3 December
- 12-14 Januari
- 16-18 Februari
Note that some of the teams may need to plan a preparation session before the first sprint, e.g., to arrange access to certain hardware or datasets.