Predictive monitoring of business processes is a challenging topic of process mining which is concerned with the prediction of process indicators of running process instances. The main value of predictive monitoring is to provide information in order to take proactive and corrective actions to improve process performance and mitigate risks in real time. In this paper, we present an approach for predictive monitoring based on the use of evolutionary algorithms. Our method provides a novel event window-based encoding and generates a set of decision rules for the run-time prediction of process indicators according to event log properties. These rules can be interpreted by users to extract further insight of the business processes while keeping a high level of accuracy. Furthermore, a full software stack consisting of a tool to support the training phase and a framework that enables the integration of run-time predictions with business process management systems, has been developed. Obtained results show the validity of our proposal for two large real-life datasets: BPI Challenge 2013 and IT Department of Andalusian Health Service (SAS).
The problem of protein structure prediction (PSP) is one of the main challenges in structural bioinformatics. To tackle this problem, PSP can be divided into several subproblems. One of these subproblems is the prediction of disulfide bonds. The disulfide connectivity prediction problem consists in identifying which nonadjacent cysteines would be cross-linked from all possible candidates. Determining the disulfide bond connectivity between the cysteines of a protein is desirable as a previous step of the 3D PSP, as the protein conformational search space is highly reduced. The most representative soft computing approaches for the disulfide bonds connectivity prediction problem of the last decade are summarized in this paper. Certain aspects, such as the different methodologies based on soft computing approaches (artificial neural network or support vector machine) or features of the algorithms, are used for the classification of these methods.
Protein structure prediction is currently one of the main open challenges in Bioinformatics. The protein contact map is an useful, and commonly used, representation for protein 3D structure and represents binary proximities (contact or non-contact) between each pair of amino acids of a protein. In this work, we propose a multiobjective evolutionary approach for contact map prediction based on physico-chemical properties of amino acids. The evolutionary algorithm produces a set of decision rules that identifies contacts between amino acids. The rules obtained by the algorithm impose a set of conditions based on amino acid properties to predict contacts. We present results obtained by our approach on four different protein data sets. A statistical study was also performed to extract valid conclusions from the set of prediction rules generated by our algorithm. Results obtained confirm the validity of our proposal.
This paper reports our experience in flipping a second-year undergraduate course on software architecture and integration, taught in the second course of a Software Engineering degree. We compare the application of the flippedclassroom methodology with a traditional methodology. Our study encompasses two academic courses, in the years 2017 and 2018, and involves a total number of 434 students and 6 lecturers, placing this among the largest studies on flipped-classroom to date. The paper also reports on the production of the videos used with the flipped-classroom methodology, recorded by the lecturers in informal settings, and provides several lessons learned in this regard. The results of the study, backed by a solid statistical analysis of the data, demonstrate the suitability of the flipped-classroom methodology for laboratory sessions in the subject course. Among other results, our analysis concluded that students had on average 24 more minutes per session to solve in-class exercises with the flipped-classroom methodology; more than 70% of the students considered that the quantity, duration and didactic content of the videos were (very) appropriate; and 9 out of every 10 students would prefer this methodology in the laboratory sessions of future courses rather than a traditional face-to-face approach.
In this paper, we focus on protein contact map prediction, one of the most important intermediate steps of the protein folding problem. The objective of this research is to know how short-range interactions can contribute to a system based on decision trees to learn about the correlation among the covalent structures of a protein residues. We propose a solution to predict protein contact maps that combines the use of decision trees with a new input codification for short-range interactions. The method's performance was very satisfactory, improving the accuracy instead using all information of the protein sequence. For a globulin data set the method can predict contacts with a maximal accuracy of 43%. The presented predictive model illustrates that short-range interactions play the predominant role in determining protein structure.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.