Tutorials


Weka Data Mining Tool

This full-day tutorial will provide an introduction to machine learning techniques for data mining based on the Java-based open-source Weka machine learning workbench developed at the University of Waikato. The tutorial will be hosted in one of the Windows computer labs at Waikato but participants can bring their own laptop, install Weka, and use it for the tutorial. During the tutorial, participants will work through parts of the on-line course “Data Mining with Weka” (https://weka.waikato.ac.nz/explorer), which provides mini-lectures and activities illustrating basic concepts in machine learning and their application in Weka. Weka experts will be at hand in the lab to answer any questions and help participants if they encounter difficulties. Participation does not require any prior knowledge in machine learning and does not involve any programming. Participants who have prior knowledge of Weka can choose to work on one of the more advanced courses “More Data Mining with Weka” or “Advanced Data Mining with Weka”.

Schedule

Full Day

Goals

To empower you to analyse your own data using a wide variety of data mining techniques — and, most importantly, to understand what you are doing! Weka is a free, open source workbench that runs on any computer and is trivial to download and install.

Target Participants

Anyone with data who wants to get to know it better.

Prerequisites

None. No laptop required — but participants can bring their own, install Weka, and use it for the tutorial.


Applying Model Driven Engineering Technologies in the Creation of Domain Specific Modeling Languages - Half day - CANCELLED


SOUFFLÉ: Datalog Compiler for Static Program Analysis - Half day

SOUFFLÉ is an open-source translator that compiles programs expressed in a Datalog-language to highly efficient parallel C++ code. Currently, SOUFFLÉ is being used for static program analysis on very large code bases. This includes points-to analysis on OpenJDK7, where it showcases its efficiency by completing analysis in under a minute. Oracle Labs deploys SOUFFLÉ for Java security analyses, since it combines the highly flexible Datalog-based program analysis paradigm with performance that is on-par with manually developed state-of-the-art tools. The first part of this tutorial will introduce the use of the specification language and tools that make up SOUFFLÉ. The language, which includes standard features of Datalog, as well as extensions (e.g., components, records, aggregates, and a strong type system), will be used in hands-on examples. The second part will cover the theory and design behind SOUFFLÉ and will give some guidance on how to participate in SOUFFLÉ’s open source community.

Schedule

Half Day

Requirements

Participants will need to bring their own laptop.

Contact

bernhard.scholz@sydney.edu.au

Interaction Design for Specifying Requirements - Half day

This tutorial has the primary objective to address a potential separation of requirements engineering and interaction design. In order to improve the development of useful and usable software systems, participants will learn about an approach to precisely specify use cases in terms of interaction design. As a positive ‘side-effect’, supporting user interfaces can be generated semi-automatically. The assumed attendee background is some familiarity with scenarios / use cases as well as interest in requirements and interaction design. There are no pre-requisites such as knowledge about human-computer interaction or requirements engineering in general. Also the small selection of UML notation used will be explained.

https://www.ict.tuwien.ac.at/en/users/kaindl

Schedule

Half Day

Learn to Build and Apply Software Analysis Tools with Interactive Visualization - Full day - CANCELLED


Dynamic analysis of JavaScript with Jalangi - Half day

JavaScript is one of the most prevalent programming languages in use today, creating a critical need for better JavaScript tools and analyses. Jalangi is a simple yet powerful framework for writing JavaScript dynamic analyses running on both modern browsers and node.js. The framework is open-source and available at:
https://github.com/Samsung/jalangi2
This tutorial will show you how to make full use of Jalangi's power, via a mix of background material and detailed walkthroughs of Jalangi analyses. We will present a detailed guide to writing Jalangi analyses, including patterns to follow, pitfalls to avoid, and debugging strategies. We will also present the MemInsight, DLint, and JITProf frameworks, all built atop Jalangi, and show how one can easily extend these frameworks to implement new rules or analyses.

Schedule

Half Day

Targeted participants

Anyone interested in dynamic analysis for JavaScript.

Pre-requisite knowledge / requirements

None. If desired, you can bring your laptop to run some Jalangi analyses on your own machine.

Contact

manu@sridharan.net

Important Dates

Tutorial Proposals:31 July 2016
Acceptance Notification:30 August 2016
Tutorials:6 December 2016