Batch-based active learning: Application to social media data for crisis management. (1st March 2018)

• Record Type: Journal Article
• Authors: Pohl, Daniela; Bouchachia, Abdelhamid; Hellwagner, Hermann
• Is Part Of: Expert systems with applications. Volume 93(2018)
• Page Start: 232
• Page End: 244
• Publication Date: 2018-03-01
• Subjects: Online learning -- Active learning -- Classification -- Social media -- Crisis management; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2017.10.026
• Languages: English
• ISSNs: 0957-4174

Abstract

Highlights: Boundary-based selection criterion to identify uncertain data instances. Uncertain data instances will be queried to enhance the accuracy of the classifier. Different selection strategies to find the instances to be labelled. Extensive experimental evaluation using different datasets. Abstract: Classification of evolving data streams is a challenging task, which is suitably tackled with online learning approaches. Data is processed instantly requiring the learning machinery to (self-)adapt by adjusting its model. However for high velocity streams, it is usually difficult to obtain labeled samples to train the classification model. Hence, we propose a novelo nlineb atch-baseda ctivel earning algorithm (OBAL) to perform the labeling. OBAL is developed for crisis management applications where data streams are generated by the social media community. OBAL is applied to discriminate relevant from irrelevant social media items. An emergency management user will be interactively queried to label chosen items. OBAL exploits the boundary items for which it is highly uncertain about their class and makes use of two classifiers: k-Nearest Neighbors (kNN) and Support Vector Machine (SVM). OBAL is equipped with a labeling budget and a set of uncertainty strategies to identify the items for labeling. An extensive analysis is carried out to show OBAL's performance, the sensitivity of its parameters, and the contribution of the individual uncertainty strategies. Two types of datasets are used: synthetic and social media datasets related to crises. The empirical results illustrate that OBAL has a very good discrimination power.