Date, Time, & Location

Tuesday, May 5, 4:00pm-5:50pm, Online

Overview

The final exam is comprehensive and will cover all material from the beginning of the semester through the end but with some focus on material covered since Test 2. The material will cover the assigned readings and the topics we discussed in class.

Format

  • Multiple Choice (20-25)
  • Free Response (5-6 questions)
  • CS680 Students will have additional questions

Topics

  • Python
  • numpy
  • pandas
  • Data (items, attributes, attribute types, semantics, metadata)
  • Data Wrangling
  • Data Cleaning
  • Data Transformation
  • Data Integration
  • Data Fusion
  • Data Exploration
  • Dataset Search
  • Scalable Databases
  • Data Curation
  • Graph Data
  • Time Series Data
  • Spatial Data
  • Provenance (Computational, Database, Evolution)
  • Reproducibility
  • Databases and Machine Learning

Readings

Assigned Readings

Referenced Papers

Free Response Example Questions

  • Examples from Test 1
  • Examples from Test 2
  • What feature of many spatial datasets do Nanocubes and TopKube take advantage of to reduce the size of the stored data?
  • What is the main difference between the types of analyses that Nanocubes and TopKube support? How does that change their implementations?
  • In addition to doing pre-computation, how does ForeCache reduce interaction latency?
  • What is provenance, and what is required to capture, store, and use provenance?
  • What is the difference between prospective and retrospective provenance?
  • What are the trade-offs between workflow- and OS-based provenance capture?
  • What questions can database provenance answer? What are the differences between “Why”, “How”, and “Where” provenance?
  • What was evolution provenance in VisTrails used for?
  • What are concerns involved in reproducing a previous computational study?
  • What types of analyses can be done to evaluate how reproducible published work is?
  • How might machine learning impact databases?
  • Which type of engine (OLAP or OLTP) is SageDB being developed for? Which components of a database does SageDB present machine learning approaches for? How do they perform versus standard databases.