Commonsense for Machine Intelligence
CIKM 2017 tutorial
A half-day tutorial, held at CIKM 2017 on November 10th.
- Beginners are introduced to commonsense knowledge mining.
- Developers can learn applications of commonsense.
- Researchers can understand the state-of-the-art and open problems.
Part 1: Acquiring commonsense knowledge [slides]
Introduction to commonsense
We introduce and compare commonsense against encyclopedic knowledge. We enumerate the types of commonsense and existing knowledge bases for each type of commonsense.
We discuss unstructured, structured, and continuous representations of commonsense knowledge.
Commonsense fact acquisition
We discuss the wide spectrum of models, including curated, distantly supervised and unsupervised models for mining commonsense from text and multimodal resources.
Evaluation of commonsense knowledge
In addition to individual facts, we discuss models to mine entailments for commonsense reasoning.
Part 2: Detecting and correcting odd collocations in text [slides]
Collocations and odd collocations
We introduce collocated expressions and odd collocations that are collocation errors where expressions are found that are not typically used in correct communication, e.g., mighty tea instead of strong tea.
Broad overview of techniques to fix collocation errors
We discuss interesting tasks where collocation error correction is useful and common techniques to fix these .
We discuss current research in the area of classifying collocation errors which leads to their correction, mainly from a linguistic classification perspective.
Detection and correction
We discuss frequency-based, semantic similarity, ranking and ensemble learning based techniques.
Part 3: Applications and open issues [slides]
We describe the primary directions in smart cities research and highlight how commonsense has been useful currently, and envision potential application of commonsense in smart cities.