-
- Framework: add more references.
-
- Background: Provide example of papers which appear to target the same problem but don't.
-
- Come up with a more interesting name than ad-eval. Anomaly detection researcher?
-
- Background: clean up Problem Definition.
-
- All: Settle on British or American english.
-
- All: Replace tasks with factor choices.
-
- Framework: Clean up Dataset Format section.
-
- Framework: Clean up Solution Format section.
-
- Time series: Rewrite implementation section.
-
- Framework: Merge Training Data and Anomaly Types into a single factor.
-
- Time series: Incorporate Important tasks section into other sections.
-
- All: Use vector notation for sequences everywhere.
-
- Framework: Make sure filters are defined the same way as in the time series chapter.
-
- All: Fix formatting issues caused by CSC:s LaTeX template.
-
- Results: Rewrite it.
-
- Conclusions: Rewrite it.
-
- Finish project plan [ABANDONED].
-
- Change name to "A Framework for Anomaly Detection with Applications to Machine-Generated Data"
-
- Reshape chapters to Introduction, Background, Framework, Application, Results, Discussion
-
- Incorporate Foreword into Introduction
-
- Capitalize chapter, section, figure, and table
-
- Move method presentation from Background to Framework
-
- Remove information on project progression on Background
-
- Add draft of formal problem definition to Background
-
- Remove Splunk-specific discussion from Background
-
- Move section 1.2 -> 2.1
-
- Framework: be more clear when introducing novel concepts.
-
- Reformulate the framework with the search problem in mind
-
- Framework: Clarify semi-supervised/supervised convention - add citation.
-
- Framework: Rename reference data -> (labelled) training data.
-
- Time series: Remove sequence extraction section.
-
- Time series: Add more formal section on sequence extraction [ABANDONED].
-
- Fix lint errors
-
- Rewrite the first half of Background to more clearly explain and motivate the work
-
- Add survey references to the Anomaly Detection section
-
- Remove rough edges resulting from the incorporation of Foreword into Introduction
-
- Introduction: update to reflect new chapter structure.
-
- Introduction: update chapter references.
-
- Rename "The Component Framework" to "A Framework for Anomaly Detection".
-
- Change "tasks" chapter name to "framework".
-
- Talk about distinction between problems and methods in Background.
-
- Rewrite the Framework introduction.
-
- All: Replace "data set" with "dataset".
-
- All: Rename "Data format" -> "Dataset format".
-
- Framework: Rename "Output format" -> "Solution format".
-
- Framework: Rename "Reference data" -> "Training data".
-
- Framework: Move last paragraph of description.
-
- Framework: Create draft of Dataset Format section.
-
- Framework: Finish discretisation discussion in Dataset Format.
-
- Framework: Add discussion of numerosity reduction to Dataset Format.
-
- Framework: Add discussion of dimensionality reduction to Dataset Format.
-
- Framework: Create draft of Solution Format section.
-
- Framework: Create draft of Training Data section.
-
- Framework: Create draft of Anomaly Type section.
-
- Framework: Create draft of Anomaly Measures section.
-
- Framework: Incorporate Related High-level Tasks section into other sections.
-
- Framework: Be more clear about context functions being a novel concept.
-
- All: Join the chapters on time series.
-
- Time series: Write an introduction.
-
- Time series: Replace Presentation section with a shorter section introducing time series.
-
- Time series: Add section summarising existing research.
-
- Time series: Rewrite oracle section.
-
- Time series: Rewrite evaluation section.
-
- Framework/Time series: Move dimensionality discussion to Framework.
-
- Abstract: Rewrite it.
-
- All: Use the CSC LaTeX template.
-
- Discussion: Rename to Conclusions.
-
- Conclusions: Rename Post-Mortem to Summary.
-
- Background: Rewrite it.