Program and Registration—2021

Welcome to AAAI-MAKE 2021/2020 – an international and virtual AI symposium. Please note that the 2021 symposium on March 22-25 has been extended by one day to include presentations from the deferred 2020 symposium and to allow attendees from different time zones (UTC -8 to +9) to participate. You will find the program and the respective start times broken down by time zones on this page. This international AI event will occur via the Zoom conference platform and can only be attended after prior registration.

Presentation Format

Symposium contributors will present virtually (see Virtual Conference Participation) their paper in a short 10-minute presentation with slides shown via the shared screen, followed by a 5-minute discussion.

Three to four presentations are grouped into sessions chaired by a member of the organizing committee or invited session chairs.

Start Times by Time Zone

Monday, March 22

Opening

Combining Machine Learning and Knowledge Engineering 📄paper
Andreas Martin and Knut Hinkelmann, FHNW

Session A

Chair: Andreas Martin 15 minutes per paper

Decision Support combining Machine Learning, Knowledge Representation and Case-Based Reasoning 📄paper
Carlo Mehli, Knut Hinkelmann and Stephan Jüngling

Supporting Case-Based Reasoning with Neural Networks: An Illustration for Case Adaptation 📄paper
David Leake, Xiaomeng Ye and David Crandall

The Significance of Computational Performance for Enterprise AI in Question-Answering 📄paper 📹recoding
Jan Meiswinkel, Patrick Levi and Rainer Hoch

Session B

Chair: Aurona Gerber 15 minutes per paper

Knowledge Infused Policy Gradients for Adaptive Pandemic Control 📄paper
Kaushik Roy, Qi Zhang, Manas Gaur and Amit Sheth

Evaluation of Machine Learning Algorithms in a Human-Computer Hybrid Record Linkage System 📄paper 📹recoding
Mahin Ramezani, Hye-Chung Kum and Guru Ilangovan

Enhancement of Record Linkage by Using Attributes containing Natural Language Text 📄paper
Philipp Behnen, Felix Kruse and Jorge Marx Gómez

Session C

Chair: Raymond Mooney 15 minutes per paper

Towards More Robust Fashion Recognition by Combining of Deep-Learning-Based Detection with Semantic Reasoning 📄paper
Achim Reiz, Mohamad Albadawi, Kurt Sandkuhl, Matthias Vahl and Dennis Sidin

Neural-Symbolic Integration for Fairness in AI 📄paper
Benedikt Wagner and Artur D’Avila Garcez

A Neural-Symbolic Approach for User Mental Modeling: A step towards building exchangeable identities 📄paper
Hossein Rahnama, Marjan Alirezaie and Alex Pentland

Session D

Chair: Hans-Friedrich Witschel 15 minutes per paper

Towards Kinematically Constrained Real Time Human Pose Estimation using Sparse IMUs 📄paper 📹recoding
Deepak Nagaraj, Rhett Dobinson and Dirk Werth

Deep Learning Approaches for Forecasting Strawberry Yields and Prices Using Satellite Images and Station-Based Soil Parameters 📄paper
Mohita Chaudhary, Mohamed Sadok Gastli, Lobna Nassar and Fakhri Karray

Anomaly Detection in Railway Infrastructure 📄paper
David Morandi and Stephan Jüngling

Tuesday, March 23

Keynote

Keynote
Doug Lenat, CEO of Cycorp, Inc. and AAAI Fellow

Session E

Chair: Knut Hinkelmann 15 minutes per paper

A Recap of Early Work on Theory and Knowledge Refinement 📄paper
Raymond Mooney and Jude Shavlik

Evaluating artificial short message service campaigns through rule based multi-instance multi-label classification 📄paper
Johannes Sahlin, Håkan Sundell, Håkan Alm and Jesper Holgersson

Implications of Combining Domain Knowledge in Explainable Artificial Intelligence 📄paper
Sheikh Rabiul Islam and William Eberle

XAI for Operations in the Process Industry – Applications, Theses, and Research Directions 📄paper
Arzam Kotriwala, Benjamin Kloepper, Marcel Dix, Gayathri Gopalakrishnan, Dawid Ziobro and Andreas Potschka

Session F

Chair: Reinhard Stolle 15 minutes per paper

Guiding Parameter Estimation of Agent-Based Modeling Through Knowledge-based Function Approximation 📄paper
William Broniec, Sungeun An, Spencer Rugaber and Ashok Goel

Exploring the Hyperparameters of XGBoost Through 3D Visualizations 📄paper
Ole-Edvard Ørebæk and Marius Geitle

Using Camera-Drones and Artificial Intelligence to Automate Warehouse Inventory 📄paper
Rene Kessler, Christian Melching, Ralph Göhrs and Jorge Marx Gómez

Session G

Chair: Knut Hinkelmann 15 minutes per paper

Batch-like Online Learning for More Robust Hybrid Artificial Intelligence: Deconstruction as a Machine Learning Process 📄paper
Thomas Schmid

Improving machine learning performance using conceptual modeling 📄paper
Arturo Castellanos, Alfred Castillo, Monica Chiarini Tremblay, Roman Lukyanenko, Jeffrey Parsons and Veda C. Storey

Hybrid AI using Graph Neural Networks in Scene Classification 📄paper
Maaike de Boer and Fieke Hillerström

Wednesday, March 24

Panel

Moderation: Knut Hinkelmann

Panel Discussion
Hans-Georg Fill, University of Fribourg, Switzerland.
Aurona Gerber, University of Pretoria, South Africa.
Doug Lenat, CEO of Cycorp, Inc. and AAAI Fellow.
Andreas Martin, FHNW University of Applied Sciences and Arts Northwestern Switzerland.
Reinhard Stolle, Argo AI GmbH, München, Germany.
Frank van Harmelen, VU University, Amsterdam, Netherlands.

Session H

Chair: Hans-Georg Fill 15 minutes per paper

Fit to Measure: Reasoning about Sizes for Robust Object Recognition 📄paper 📹recoding
Agnese Chiatti, Enrico Motta, Enrico Daga and Gianluca Bardaro

ZERO – Detect objects without training examples, by knowing their parts. 📄paper
Gertjan Burghouts and Fieke Hillerström

Knowledge Engineering with Image Data in Real-World Settings (DEMO) 📄paper 📹recoding
Margaret Warren, David A. Shamma and Pat Hayes

Session I

Chair: Aurona Gerber 15 minutes per paper

EmEL++ : Embeddings for the EL++ Description Logic 📄paper
Sutapa Mondal, Sumit Bhatia and Raghava Mutharaju

Towards Automatic Ontology Alignment using BERT 📄paper
Sophie Neutel and Maaike de Boer

Ontology-based transfer learning in the airport and warehouse logistics domains 📄paper 📹recoding
Miriam Herold and Mirjam Minor

Session J

Chair: Reinhard Stolle 15 minutes per paper

Neuro-Symbolic Deductive Reasoning for Cross-Knowledge Graph Entailment 📄paper 📹recoding
Monireh Ebrahimi, Md Kamruzzaman Sarker, Federico Bianchi, Ning Xie, Aaron Eberhart, Derek Doran, Hyeongsik Kim and Pascal Hitzler

Learning profile-based recommendations for medical search auto-complete 📄paper
Guusje Boomgaard, Selene Báez Santamaría, Ilaria Tiddi, Robert-Jan Sips and Zoltán Szlávik

Using a General Prior Knowledge Graph to Improve Data-Driven Causal Network Learning 📄paper
Meghamala Sinha and Stephen Ramsey

Session K

Chair: Pascal Hitzler 15 minutes per paper

Hybrid Conversational AI for Intelligent Tutoring Systems 📄paper
Charuta Pande, Hans Friedrich Witschel, Andreas Martin and Devid Montecchiari

Semantic Enrichment of Pretrained Embedding Output for Unsupervised IR 📄paper
Edmund Dervakos, Giorgos Filandrianos, Konstantinos Thomas, Alexios Mandalios, Chrysoula Zerva and Giorgos Stamou

Automated Construction of Knowledge-Bases for Safety Critical Applications: Challenges and Opportunities 📄paper
Amit Bhatia and Alessandro Pinto

Thursday, March 25 (extended)

Session L

Chair: Aurona Gerber 15 minutes per paper

The FATE system: FAir, Transparent and Explainable decision making 📄paper
Joachim de Greeff, Maaike de Boer, Fieke Hillerström, Freek Bomhof, Wiard Jorritsma and Mark Neerincx

SAMS: Human-in-the-loop Approach to Combat the Sharing of Digital Misinformation 📄paper
Shaban Shabani, Zarina Charlesworth, Maria Sokhn and Heiko Schuldt

Team Design Patterns for Moral Decisions in Hybrid Intelligent Systems: A Case Study of Bias Mitigation 📄paper 📹recoding
Jip van Stijn, Mark Neerincx, Annette ten Teije and Steven Vethman

Session M

Chair: Hans-Georg Fill 15 minutes per paper

“Is depression related to cannabis?”: A Knowledge-Infused model for Entity and Relation Extraction with Limited Supervision 📄paper
Kaushik Roy, Usha Lokala, Vedant Khandelwal and Amit Sheth

Towards unsupervised knowledge extraction 📄paper
Dorothea Tsatsou, Konstantinos Karageorgos, Anastasios Dimou, Javier Carbo, Jose Molina and Petros Daras

Capturing Knowledge of Emerging Entities from Extended Search Snippets 📄paper
Sunday Ngwobia, Saeedeh Shekarpour and Faisal Alshargi

Session N

Chair: Andreas Martin 15 minutes per paper

Case-Based System of User Interface Reverse Engineering 📄paper
Pavel Myznikov

A Study in Practical Solutions to Sarcasm Detection with Machine Learning and Knowledge Engineering Techniques 📄paper
Chia Zheng Lin, Michal Ptaszynski, Masui Fumito, Gniewosz Leliwa, Michal Wroczynski

Combining Symbolic and Sub-symbolic AI in the Context of Education and Learning 📄paper 📹recoding
Rainer Telesko, Stephan Jüngling, Phillip Gachnang

Session O

Chair: Knut Hinkelmann 15 minutes per paper

Common-Knowledge Concept Recognition for SEVA 📄paper
Jitin Krishnan, Patrick Coronado, Hemant Purohit, Huzefa Rangwala

Attribution-based Salience Method towards Interpretable Reinforcement Learning 📄paper
Yuyao Wang, Masayoshi Mase, Masashi Egi

Using Pre-trained Transformer Deep Learning Models to Identify Named Entities and Syntactic Relations for Clinical Protocol Analysis 📄paper
Miao Chen, Fang Du, Ganhui Lan, Victor Lobanov

Closing

Concluding AAAI-MAKE
Knut Hinkelmann and Andreas Martin, FHNW

Virtual Conference Participation

All participants will receive credentials for accessing the virtual symposium meeting. This information will be sent to you shortly before the symposium via the email address you used for registration.

The virtual symposium meeting will be hosted on the Zoom conference platform. Please note to have Zoom (at least version 5.0) installed (https://zoom.us/download) in advance of the symposium.

It is appreciated that participants mute their microphones when not speaking but activate their webcam throughout the symposium.

Video Recording and Release

During the AAAI-MAKE meeting, it is intended to record selected presentations and make them available online later. The recording will be configured so that only the active speaker’s video with the shared screen will be recorded exclusively.

If presenters wish a recording of their presentation, the AAAI photograph release form must be filled out, signed and uploaded, before the symposium.

Registration*

Accepted authors, invited speakers, symposium participants, and other invited attendees must register to participate at the virtual symposia. Please register via the online registration form (https://aaaiconf.cventevents.com/sss21).

* The Association for the Advancement of Artificial Intelligence (AAAI) is responsible for the overall organization of all the Spring Symposium Series. For any inquiries related to registration please contact the AAAI at sss21@aaai.org or consult the AAAI’s website (https://aaai.org/Symposia/Spring/sss21.php).