VIII Workshop on Data and Knowledge Engineering

WDKE-2024, September 4, 2024

Part of The XIV International Conference on Computing and Informatics in the North of Chile, INFONOR CHILE 2024

Organized by Research Core in Artificial Intelligence and Data Science, Universidad Católica del Norte, Antofagasta, Chile.

Sponsored by Master of Science in Informatics Engineering, Universidad Católica del Norte

Workshop Chairs: Claudio Meneses [cmeneses@ucn.cl], Brian Keith [brian.keith@ucn.cl], Francisco García [francgar@unap.cl] and Víctor Flores [vflores@ucn.cl]

Link for videconference (short presentations and Invited Talk #3)

Workshop on Data and Knowledge Engineering

The Workshop on Data and Knowledge Engineering (WDKE) is a space for the dissemination of scientific, professional and academic activity in the area of data and knowledge enginnering, including:

• Invited talks
• Presentations of short papers

Topics of Interest

WDKE accepts research works on the subject of data and knowledge engineering, mainly on but not limited to:

• Data Science
• Big Data
• Intelligent systems
• Data Analytics
• Information Visualization
• Web/Data mining
• Machine Learning
• Deep Learning
• Knowledge Representation
• Expert Systems
• Generative Artificial Intelligence
• Large Language Models
• Recommender Systems
• Computer Vision

Important Dates (extended deadlines):

• Submission Deadline: July 14, July 28, 2024
• Author Notification: August 5 August 13, 2024
• Camera Ready: August 21 August 28, 2024

Paper Submission

• Authors are invited to submitt short papers of 5-9 pages (including references) that report results of scientific research. All papers must be formatted with the CEURART single-column style, available at https://ceur-ws.org/Vol-XXX/CEURART.zip
• Papers must be written in English and should be submitted through the Easychair platform, visiting the following URL: Easychair-WDKE2024.
• The submitted papers will be reviewed by at least two members of an international program committee. WDKE2024 Proceedings shall be submitted to CEUR-WS.org for online publication, which is indexed by SCOPUS.

Author Registration

• At least one author from each accepted paper must be registered for the INFONOR CHILE Conference INFONOR-2024 by the 29th of August, 2024.

Final General Programme

Detailed Programme

Presentations of short papers

Link for videoconference

• September 4, 2024, 16.30 - 18.10 hrs.[Chile Continental Time; GMT-4]
• Room: Auditorium Karina Carvajal, 2nd floor, DIICC, Universidad de Atacama, Copiapó, Chile

  Time	Activity
  16.30-16.50	Paper 1: Víctor Flores, Ingrid Bravo and Rafael Martinez. Assessment of River Water Quality Using Open-Access Physicochemical Datasets and Machine Learning: A Study of the Loa River in the Atacama Desert (Chile)
  16.50-17.10	Paper 2: Vicky Rivero, Claudio Meneses and Jaime A. Pavlich-Mariscal. A semi-automatic process to resolve anaphoric ambiguities in software requirements
  17.10-17.30	Paper 3: José Benítez, Víctor Flores, Rafael Martinez and Adel Elmaghraby. Optimization of copper recovery by Flotation using Machine Learning
  17.30-17.50	Paper 4: Jorge Rivera, Scarlett Zapata, Ricardo Pizarro and Brian Keith. Enhancing Chatbot Performance with Retrieval Augmented Generation and Prompt Engineering
  17.50-18.10	Paper 5: Yashwanth Reddy Kistipati, Raja Sekhar Reddy Vangala, Ajay Kumar Reddy Padedam and Sai Nithin Sheri. Enhancing Customer Experience Through Multi Modal Sentiment Analysis
  -	-

Invited Talk #3: Title: Regression Analysis When Both Variables are in Error.

Link for videoconference

• September 4, 2024, 18.15- 18.55 hrs.[Chile Continental Time; GMT-4]

• Room: Auditorium Karina Carvajal, 2nd floor, DIICC, Universidad de Atacama, Copiapó, Chile

• Speaker:Dr. Ranjit Das

• Affiliation:Universidad Católica del Norte, Antofagasta, Chile

• Abstract: Orthogonal regression is commonly used to correct for measurement errors in predictors. However, its application can lead to overestimation of slopes due to the often-overlooked equation errors, resulting in excessive correction for measurement error. In this lecture, I will address the pitfalls of using orthogonal regression in the context of errors-in-variables linear regression, highlighting the need for a thorough assessment of equation errors rather than relying solely on informal estimates of measurement error variances. I will introduce Advanced Orthogonal Regression (AOR), a refined method that improves the accuracy of conventional orthogonal regression by converting independent variables into dependent variables more effectively. Traditional orthogonal regression establishes a linear relationship between the dependent variable (Yt) and the independent variable (Xt) based on observed data (Xobs , Yobs ). However, using observed data directly in the equation can introduce bias in estimating Yt due to errors in Xobs . I will demonstrate how utilizing Xt —derived through standard linear regression (SLR) between Yt and Xobs — provides more accurate estimates and reduces this bias. This lecture will offer insights into the limitations of conventional orthogonal regression and present the advantages of the AOR method, providing valuable techniques for data scientists working with measurement errors in predictive modeling.**

• Short Bio: Dr. Ranjit Das is an internationally recognized researcher and industry expert in Earthquake Engineering. He has developed a new earthquake magnitude scale and created several fundamental algorithms for data analysis using regression techniques. As a UNESCO-invited scientist specializing in Earthquake and Tsunami Hazards, Dr. Das serves as an Associate Editor for the Journal of Geophysics, the world oldest geophysical journal, and recently in the Scientific Report journal. He is also a member of the editorial board for the Discover Civil Engineering Journal (Springer Nature), editor of the Bioinformatics and Computational Biology journal, and a member of Sigma Xi. At Universidad Católica del Norte, Chile, where he is an Assistant Professor, Dr. Das focuses on Regression Analysis, Seismic Hazard and Risk Assessment, and Statistical Seismology. He earned his Ph.D. in Statistical Seismology from the Indian Institute of Technology Roorkee, India. His extensive research includes 26 peer-reviewed high-impact scientific publications and contributions to widely used algorithms in science and engineering. Dr. Das is a life member of the Seismological Society of America

Technical Program Committee