AAAI 2019 Spring Symposium on Combining Machine Learning with Knowledge Engineering (AAAI-MAKE)

Symposium Program and Papers

March 25–27, 2019 @ Stanford University, Palo Alto, California, USA

Monday, March 25

Keynote & Opening Session

Symposium Opening • Andreas Martin (Slides) and Knut Hinkelmann (Slides), FHNW

9:00 - 9:30

Keynote: It’s Statistical Learning and Knowledge Engineering All the Way Down Doug Lenat, CEO of Cycorp, Inc. and AAAI Fellow

9:30 - 10:15

Coffee Break

10:30 - 11:00

Position Paper Session

Chair: Aurona Gerber11:00 - 12:30 15-20 minutes per paper

Deconstructing the Final Frontier of Artificial Intelligence: Five Theses for a Constructivist Machine LearningThomas Schmid

Knowledge engineering and machine learning for design and use in cyber-physical environmentsMichael Walch

From Demonstrations and Knowledge Engineering to a DNN Agent in a Modern Open-World Video Game • (Slides) Igor Borovikov and Ahmad Beirami

Leverage White-Collar Workers with AI • (Slides) Stephan Jüngling and Angelin Hofer


12:30 - 14:00

Full Paper Session

Chair: Doug Lenat14:00 - 15:30 15-20 minutes per paper

SEVA: A Systems Engineer’s Virtual AssistantJitin Krishnan, Patrick Coronado and Trevor Reed

Virtual Bartender: A Dialog System Combining Data-Driven and Knowledge-Based Recommendation • (Slides) Knut Hinkelmann, Monika Blaser, Oliver Faust, Alexander Horst and Carlo Mehli

Building Knowledge Base through Deep Learning Relation Extraction and WikidataPero Subasic, Hongfeng Yin and Xiao Lin

Coffee Break

15:30 - 16:00

Position & Full Paper Session

Chair: Knut Hinkelmann16:00 - 17:30 15-20 minutes per paper

Collaborative Learning of Concept Representations for Video DataFrancisco Torres, Hoda Eldardiry, Gaurang Gavai and Chad Ramos

Representation and Retrieval of Images by Means of Spatial Relations Between Objects • (Slides) Danilo Nunes, Leonardo Anjoletto Ferreira, Paulo E. Santos and Adam Pease

Modularity as a Means for Complexity Management in Neural Networks LearningDavid Castillo-Bolado, Cayetano Guerra-Artal and Mario Hernandez-Tejera

Day 1 Concluding • Frank van Harmelen, VU University

17:10 - 17:30


18:00 - 19:00

Tuesday, March 26

Keynote Session

Keynote • (Slides) Frank van Harmelen, VU University

9:00 - 09:40

Position & Full Paper Session

Chair: Andreas Martin9:40 - 10:30 15-20 minutes per paper

Applying the Concept of Knowledge Blockchains to OntologiesHans-Georg Fill

Combining Machine Learning with Knowledge Engineering to detect Fake News in Social Networks - A Survey • (Slides) Knut Hinkelmann, Sajjad Ahmed and Flavio Corradini

Coffee Break

10:30 - 11:00

Position & Full Paper Session

Chair: Doug Lenat11:00 - 12:30 15-20 minutes per paper

Improving Topic Modeling for Textual Content with Knowledge Graph EmbeddingsMarco Brambilla and Birant Altinel

The Lokahi Prototype: Toward the automatic Extraction of Entity Relationship Models from Text • (Slides) Michael Kaufmann

DEXTER - Data EXTraction & Entity Recognition for Low Resource DatasetsNihal V. Nayak, Pratheek Mahishi and Sagar Rao

CoKE : Word sense induction using contextualized knowledge embeddingsSanjana Ramprasad and James Maddox


12:30 - 14:00

Poster & Demonstration Session

Chair: Knut Hinkelmann

PLOW: Probabilistic Logic Over the Well-Founded Semantics • (Slides) Benjamin Grosof and Theresa Swift

Poster and Demonstration14:00 - 14:20

Using Predicate Information from a Knowledge Graph to Identify Disease Trajectories • (Slides) Wytze Vlietstra, Rein Vos, Erik van Mulligen and Jan Kors

Poster14:20 - 14:35

A proposal for determining the evidence types of biomedical documents using a drug-drug interaction ontology and machine learningLinh Hoang, Richard D. Boyce, Mathias Brochhausen, Joseph Utecht and Jodi Schneider

Poster14:35 - 14:50

On Datasets for Evaluating Architectures for Learning to ReasonNaveen Sundar Govindarajulu, Jean-Claude Paquin, Shreya Banerjee, Atriya Sen, Paul Mayol and Selmer Bringsjord

Poster14:50 - 15:05

Learning Temporal Rules from Knowledge Graph StreamsPouya Ghiasnezhad Omran, Kewen Wang and Zhe Wang

Poster and Full Paper15:05 - 15:20

Coffee Break

15:30 - 16:00

Full Paper Session

Chair: Frank van Harmelen16:00 - 17:40 15-20 minutes per paper

Duplicate Removal for Overlapping Clusters: A Study Using Social Media Data • (Slides) Amit Paul and Animesh Dutta

Performance measures fusion for experimental comparison of methods for multi-label classificationTome Eftimov and Dragi Kocev

CrystalGAN: Learning to Discover Crystallographic Structures with Generative Adversarial NetworksAsma Nouira, Nataliya Sokolovska and Jean-Claude Crivello

Tree-based Regularization for Interpretable Readmission PredictionJialiang Jiang, Sharon Hewner and Varun Chandola

Day 2 Concluding • Knut Hinkelmann, FHNW

17:20 - 17:40

Plenary Session (Slides) Aurona Gerber, University of Pretoria, reporting in plenary session

18:00 - 19:00

Wednesday, March 27

Position & Full Paper Session

Chair: Frank van Harmelen9:00 - 10:30 15-20 minutes per paper

Towards an Assistive and Pattern Learning-driven Process Modeling Approach • (Slides) Emanuele Laurenzi, Knut Hinkelmann, Stephan Jüngling, Devid Montecchiari, Charuta Pande and Andreas Martin

DeepLogic: Towards End-to-End Differentiable Logical ReasoningNuri Cingillioglu and Alessandra Russo

On the Capabilities of Logic Tensor Networks for Deductive ReasoningFederico Bianchi and Pascal Hitzler

KINN: Incorporating Expert Knowledge in Neural Networks • (Slides)Muhammad Ali Chattha, Shoaib Ahmed Siddiqui, Muhammad Imran Malik, Ludger Van Elst, Andreas Dengel and Sheraz Ahmed

Coffee Break

10:30 - 11:00

Position & Full Paper Session

Chair: Aurona Gerber11:00 - 12:30 15-20 minutes per paper

Learning and Engineering Similarity Functions for Business Recommenders and Clustering • (Slides) Hans Friedrich Witschel and Andreas Martin

Leveraging Wikipedia for Ontology Pattern Population from Text • (Slides) Michelle Cheatham, James Lambert and Charles Vardeman II

Metrics for Evaluating Quality of Embeddings for Ontological ConceptsFaisal Alshargi, Saeedeh Shekarpour, Tommaso Soru and Amit Sheth

Day 3 and Symposium Concluding • Knut Hinkelmann and Andreas Martin, FHNW

12:00 - 12:30