A Rich Feature-based Kernel Approach for Drug- Drug Interaction Extraction
Journal Title: International Journal of Advanced Computer Science & Applications - Year 2017, Vol 8, Issue 4
Abstract
Discovering drug-drug interactions (DDIs) is a crucial issue for both patient safety and health care cost control. Developing text mining techniques for identifying DDIs has attracted a great deal of attention in the last few years. Unfortunately, state-of-the-art results didn't exceed the threshold of 0.7 F1 score, which calls for more efforts. In this work, we propose a new feature-based kernel method to extract and classify DDIs. Our approach consists of two steps: identifying DDIs and assigning one of four different DDI types to the predicted drug pairs. We demonstrate that by using new groups of features non-linear kernels can achieve the best performance. When evaluated on the DDIExtraction 2013 challenge corpus, our system achieved an F1-score of 71.79%, as compared to 69.75% and 68.4% reported by the top two state-of-the-art systems.
Authors and Affiliations
ANASS RAIHANI, NABIL LAACHFOUBI
Keywords
Related Articles
- EP ID EP258359
- DOI 10.14569/IJACSA.2017.080445
- Views 91
- Downloads 0
Nonlinear Condition Tolerancing Using Monte Carlo Simulation
To ensure accuracy and performance of the products, designers tend to hug the tolerances. While, manufacturers prefer to increase them in order to reduce costs and ensure competition. The analysis and synthesis of tolera...
From PID to Nonlinear State Error Feedback Controller
In this paper an improved nonlinear state error feedback controller (INLSEF) has been proposed for perfect reference tracking and minimum control energy. It consists of a nonlinear tracking differentiator together with n...
Improving K-Means Algorithm by Grid-Density Clustering for Distributed WSN Data Stream
At recent years, Wireless Sensor Networks (WSNs) had a widespread range of applications in many fields related to military surveillance, monitoring health, observing habitat and so on. WSNs contain individual nodes that...
Ladder Networks: Learning under Massive Label Deficit
Advancement in deep unsupervised learning are finally bringing machine learning close to natural learning, which happens with as few as one labeled instance. Ladder Networks are the newest deep learning architecture that...
Location-aware Event Attendance System using QR Code and GPS Technology
Attendance process in a university’s event is time consuming and tracking the attendance can be harder. In this paper, a smart event attendance system for a university using QR code and GPS technology is proposed with ob...