The results of the survey have been published in the public SALERO Deliverable D3.1.4 which is available here.

ONTOCOM cost estimation model

Data collection for model calibration

ONTOCOM background information

In case you have any questions please contact Tobias Bürger.

STI Innsbruck SALERO

INTRODUCTION

ONTOCOM is a cost estimation model for the area of Ontology Engineering developed at the Free University of Berlin in collaboration with the Institute AIFB (University of Karlsruhe - TH). It aims at predicting the costs arising in typical classes of ontology engineering processes such as ontology building, ontology reuse or ontology maintenance.

The accuracy of the cost predictions of Ontology Engineering tasks calculated by means of ONTOCOM can be improved by the calibration of the model. This necessitates the collection of real-world ontology engineering project data. In STI Innsbruck we are accomplishing the calibration of the model for the evaluation of the efforts incurred for the creation of some ontologies and this survey serves for the collection of the required data. The calibration is part of the task T3.3 "Evaluation of the economics of multimedia ontologies" of the European project SALERO.

ABOUT THIS SURVEY

Developed in the tradition of Software Engineering, ONTOCOM uses a parametric prediction equation which contains product-, personnel and project management-related effort multipliers. The effort multipliers are used to adjust the nominal effort to reflect the particularities of the ontology and of the underlying engineering process. They are rated accordingly with values from Very Low to Very High, depending on their (positive or negative) impact on the nominal development effort.

The survey, consisting of 38 questions, is divided into 4 parts:

  • introductory questions (Questions 1 to 13)
  • product-related questions (Questions 14 to 31)
  • personnel-related questions (Questions 32 to 35)
  • project-related questions (Questions 36 to 38)

For most of the questions you are required to specify the value of a certain effort multiplier, i.e. to position your answer according to a five-step rating scale. If a particular activity induces an increase of the nominal ontology development effort, then it should be rated with values such as High and Very High. In contrast, if it causes a decrease of the nominal costs, then it should be rated with values such as Low and Very Low. Finally, if the corresponding activity does not influence the nominal costs it should be rated with Nominal. THIS APPLIES IN PARTICULAR TO ACTIVITIES WHICH DO NOT APPLY TO YOUR PROJECT SETTING. For each effort multiplier, we suggest decision criteria which could be taken into consideration when assigning the corresponding ratings.

This survey is especially tailored for multi-media ontologies.

STATUS / RESULTS

This survey is closed. We collected more than 50 multimedia ontologies. The authors of these ontologies were asked to contribute to this survey by mail. We received around 20 responses. The data was used to calibrate the Ontocom and the prediction accuracy could be improved! Thanks to everyone who contributed to this survey.

Please check back for the survey results - they will be published in a consolidated form soon.

Best regards,

Tobias Bürger