 |
The First International Conference on SOTICS 2011 |
||
| Tutorials |
T1. An Integrated Approach to ICT-based Process Innovation
Luigi Lavazza
Università degli Studi dell’Insubria - Varese // CEFRIEL - Milano, Italy
T2. Scenario-based Requirements Engineering and User-Interface Design
Hermann Kaindl
Vienna University of Technology, Austria
T3.System Identification and Data Mining with HeuristicLab
Gabriel Kronberger and Michael Kommenda
Hagenberg Upper Austria University of Applied Sciences, Austria
DETAILS
T1. An Integrated Approach to ICT-based Process Innovation
Luigi Lavazza
Università degli Studi dell’Insubria - Varese // CEFRIEL - Milano, Italy
In many cases it is necessary to evaluate ICT processes or ICT-based processes. An example is when a Public Administration needs to evaluate the outcomes of investments in the automation of administrative process, either to provide services to citizens or to improve its own efficiency. In general, a quantitative evaluation is needed, e.g., when the goal is to assess how effectively the money from taxes has been used. Similar evaluation problems can be found in companies that invest on information systems and need to prove that the results are worth the expenses.
In all these cases the evaluator faces several problems:
- to understand the process and what aspects need to be evaluated;
- to make explicit and document the objective of the evaluation;
- to prepare a measurement plan (also considering the feasibility and cost of the measurement);
- to devise a way for collecting and storing measures;
- to devise effective ways for analyzing and interpreting the collected data, and presenting the results;
- to find suitable tools that can ease the whole process;
- to let everything fit in the company's strategy.
In practice the evaluator finds that several methods and techniques are available for addressing single items of the list above, but no method and toolset is available for addressing the whole work in an integrated way. Note that the integration has to hold both at the conceptual and operational (i.e., tool-supported) levels.
The proposed tutorial addresses the problem of integrating the main aspects of process evaluation. A technique and tool (based on the GQM [1][2][3] and GQM+strategies [4] methods) are presented. A case study is performed by means of the GQM toolset, which integrates:
- A tool that supports the definition of GQM plans. Conceptually, a top-down derivation of metrics definitions from high-level goals is supported.
- A database for storing measures.
- The connection of queries to elements of the GQM plan. Via the DBMS it is possible to execute the queries and associate results to the GQM plan elements.
- A statistical tool that can perform different kinds of analyses of the measures.
- A reporting tool that collects the various types of results and provides an integrated view of the situation at the GQM goal level.
The tutorial is addressed to all those interested in process evaluation and measurement. No specific knowledge is required as a prerequisite.
References
- Basili, V. and Weiss D. (1984) “A methodology for collecting valid software engineering data” IEEE Transactions on Software Engineering, vol. SE-10, no. 6, pp. 728-738.
- Basili, V. and Rombach, D. (1988) “The TAME project: towards improvement-oriented software environments,” IEEE Transactions on Software Engineering, vol. 14, no. 6, pp. 758-773.
- Basili, V., Caldiera, G., and Rombach, D. (1994) “Goal/Question/Metric Paradigm,” in Encyclopedia of Software Engineering, vol. 1, J. C. Marciniak, Ed.: John Wiley & Sons, pp. 528-532.
- Basili, V., Lindvall, M., Regardie, M., Seaman, C., Heidrich, J., Munch, J., Rombach, D. and Trendowicz, A. (2010) “Linking Software Development and Business Strategy Through Measurement” Computer, vol. 43, no. 4, pp. 57-65.
T2. Scenario-based Requirements Engineering and User-Interface Design
Hermann Kaindl
Vienna University of Technology, Austria
When the requirements and the user-interface design of a system are separated, they will most likely not fit together, and the resulting system will be less than optimal. Even if all the real needs are covered in the requirements and also implemented, errors may be induced by human-computer interaction through a bad user interface. Such a system may even not be used at all. Alternatively, a great user interface of a system with features that are not required will not be very useful as well.
Therefore, the primary motivation of this tutorial is to improve system development in practice both regarding requirements engineering and user-interface design, especially facilitating the latter. We argue for combined requirements engineering and user-interface design, primarily based on usage scenarios in the sense of sequences of actions aimed at accomplishing some task goal. However, scenario-based approaches vary especially with regard to their use, e.g., employing abstract use cases or integrating scenarios with functions and goals in a systematic design process. So, the key issue to be addressed is how to combine different approaches, e.g., in scenario-based development, so that the result is an overall useful and useable system. In particular, scenarios are very helpful for purposes of usability as well.
Prerequisite knowledge
The assumed attendee background is primarily some interest in requirements engineering or user interfaces.
Related publications of the presenter
1. Kaindl, H. (1997) A Practical Approach to Combining Requirements Definition and Object-Oriented Analysis. Annals of Software Engineering, 3, 319-343.
2. Kaindl, H. (2000) A Design Process Based on a Model Combining Scenarios with Goals and Functions. IEEE Transactions on Systems, Man, and Cybernetics (SMC) Part A, 30, 537- 551.
3. Kaindl, H. (2001) Adoption of Requirements Engineering: Conditions for Success. Proceedings of the Fifth IEEE International Symposium on Requirements Engineering (RE'01), invited State-of the-Practice Talk, Toronto, Canada, August, pp. 156-163. IEEE.
4. Kaindl, H. (2005) Is Object-oriented Requirements Engineering of Interest? Requirements Engineering, 10, 81-84.
5. Kaindl, H. (2005) A Scenario-Based Approach for Requirements Engineering: Experience in a Telecommunication Software Development Project. Systems Engineering, 8, 197-210.
6. Kaindl, H (2009) Combining Requirements and Interaction Design through Usage Scenarios. In: Human-Computer Interaction — INTERACT 2009, Proceedings of the 12th IFIP TC 13 International Conference, Part II, LNCS 5727, Springer, 932-933.
7. Kaindl, H. and Jezek, R. (2002) From Usage Scenarios to User Interface Elements in a Few Steps. Proceedings of the Fourth International Conference on Computer-Aided Design of User Interfaces (CADUI’2002), Valenciennes, France, May, pp. 91-102. Kluwer Academic Publishers, Dordrecht, The Netherlands.
8. Kaindl, H., Kramer, S. and Hailing, M. (2001) An Interactive Guide Through a Defined Modelling Process. in People and Computers XV, Joint Proceedings of HCI 2001 and IHM 2001, Lille, France, September, pp. 107-124. Springer-Verlag, London, England.
9. Kaindl, H. and Svetinovic, D. (2010) On confusion between requirements and their representations. Requirements Engineering, 15, 307-311.
10. Mukasa, K. and Kaindl, H. (2008) An Integration of Requirements and User Interface Specifications, In Proceedings of the 16th IEEE International Requirements Engineering Conference (RE 2008), 327-328.
T3.System Identification and Data Mining with HeuristicLab
Gabriel Kronberger and Michael Kommenda
Hagenberg Upper Austria University of Applied Sciences, Austria
The proposed tutorial demonstrates how to apply HeuristicLab [1] for solving data analysis problems. It will be shown how to parameterize and execute different algorithms including linear methods, support vector machines, and genetic programming to solve data analysis problems, in particular regression, classification, and time-series prognosis.
After a brief introduction to data analysis and a discussion of frequently encountered pitfalls that must be avoided in the preparation of experiments, we will show how to use different data analysis algorithms implemented in HeuristicLab to create classification, regression, and time-series prognosis models. A major focus will be put on evolutionary system identification with genetic programming, a powerful and accessible modeling approach which is capable of identifying non-linear systems and which produces white-box models in the form of symbolic mathematical expressions or simple if-then-else rules. We will also show how to calculate the relevance of input variables for the prediction of a given target variable.
The attendees will learn the fundamentals of data-analysis algorithms for practical applications and acquire hands-on experience in using HeuristicLab to prepare and parameterize data-analysis expe-riments for optimal results and in using HeuristicLab’s graphical user interface for model simplifica-tion, analysis and knowledge discovery.
