Ontological Evolutionary Encoding to Bridge Machine Learning and Conceptual Models: Approach and Industrial Evaluation

On the influence of model fragment properties on a machine learning-based approach for feature location

“…1. Encoding: The ontology is used to encode the triplets of a training dataset into feature vectors, as described in [21]. Moreover, since both feature descriptions and model fragments are based on natural language, the terms used in the ontology do not always align well with the terms in the feature description and with the terms in the model fragments.…”

Traceability Link Recovery between Requirements and Models using an Evolutionary Algorithm Guided by a Learning to Rank Algorithm: Train control and management case

“…In [22], we proposed an encoding where each model fragment is encoded as a feature vector taking into account an ontology. Specifically, each concept and relation in the ontology is represented as a feature in the feature vector.…”

“…In addition, Figure 12 also shows a limitation of the encoding proposed in [22] when it is applied at the model-element level. Several feature vectors contain the same feature values, but different target values (e.g., both pantographs).…”

“…In [22], the authors propose an evolutionary ontological encoding approach to enable Machine Learning techniques to be used to perform Software Engineering tasks in models. Their proof of concept consists of recovering traceability links between requirements and model fragments.…”

Opportunities for software reuse in an uncertain world: From past to emerging trends

“…Reengineering firmware into SPLs was possible through feature location. These feature location efforts range from Information Retrieval to Machine Learning, and include the dimension of Search‐based Software Engineering.…”