TAILIEUCHUNG - Single concatenated input is better than indenpendent multiple input for CNNs to Predict chemical-induced disease relation from literature

This paper demonstrates that for a CNN it is vice-versa, in which concatenation is better for CDR classification. To this end, we develop a CNN based model with multiple input concatenated for CDR classification. Experimental results on the benchmark dataset demonstrate its outperformance over other recent state-of-the-art CDR classification models. | VNU Journal of Science Comp. Science amp Com. Eng Vol. 36 No. 1 2020 11-16 Original Article Single Concatenated Input is Better than Indenpendent Multiple-input for CNNs to Predict Chemical-induced Disease Relation from Literature Pham Thi Quynh Trang Bui Manh Thang Dang Thanh Hai Bingo Biomedical Informatics Lab Faculty of Information Technology VNU University of Engineering and Technology Vietnam National University Hanoi 144 Xuan Thuy Cau Giay Hanoi Vietnam Received 21 October 2019 Revised 17 March 2020 Accepted 23 March 2020 Abstract Chemical compounds drugs and diseases are among top searched keywords on the PubMed database of biomedical literature by biomedical researchers all over the world according to a study in 2009 . Working with PubMed is essential for researchers to get insights into drugs side effects chemical-induced disease relations CDR which is essential for drug safety and toxicity. It is however a catastrophic burden for them as PubMed is a huge database of unstructured texts growing steadily very fast 28 millions scientific articles currently approximately two deposited per minute . As a result biomedical text mining has been empirically demonstrated its great implications in biomedical research communities. Biomedical text has its own distinct challenging properties attracting much attetion from natural language processing communities. A large-scale study recently in 2018 showed that incorporating information into indenpendent multiple-input layers outperforms concatenating them into a single input layer for biLSTM producing better performance when compared to state-of-the-art CDR classifying models. This paper demonstrates that for a CNN it is vice-versa in which concatenation is better for CDR classification. To this end we develop a CNN based model with multiple input concatenated for CDR classification. Experimental results on the benchmark dataset demonstrate its outperformance over other recent state-of-the-art CDR classification .

• Hệ điều hành
• hemical disease relation prediction
• Convolutional neural network
• Biomedical text mining
• Chemical compounds
• Chemical induced disease relations
• Communication engineering
• Engineering drawing
• Engineering design
• Lecture Engineering drawing and design
• Design communication
• Technical writing
• Graphical communication
• communication systems
• communication systems research
• communication applications
• electronic media
• communication solutions
• systems management
• Oral presentation
• International journal of computer science & communication networks
• Study of component based software engineering
• Component software engineering
• Domain engineering and component
• Software Engineering encompasses two parallel engineering activities
• International Journal of Computer Networks and Communications Security
• Performance analysis of multimedia traffic
• MPLS communication networks with traffic engineering
• MPLS communication networks
• Conventional IP network
• Engineering Optimization
• Wireless Communication
• Communication Networks
• Swarm Intelligence
• Power Engineering
• Fuzzy Cognitive
• Maps Learning
• Project based learning
• Engineering students
• Communication skill
• Foreign language
• Engineering graduates
• Electronic learning courses
• Communication modes
• Technical drawing standards
• Mechanical engineering
• Basics of mechanical engineering
• Engineering scrutiny
• Fluid mechanics
• Written and oral communication
• Energy systems
• The nature of design
• Communication of design
• Philosophy of designing
• Software engineering
• Lecture Software engineering
• Software process
• Công nghệ phần mềm
• Software engineering knowledge
• Communication principles
• Công nghệ Kỹ thuật Điện tử
• Kỹ thuật viễn thông
• Marter's thesis in Electronics
• Posture recognitions
• Thresshold investigations
• processes of manufacture
• Technology
• assembly
• communication of engineering
• legal specification
• international rules
• Scholarly Journals
• Engineers’ Communication
• Information Output
• Engineering Education
• Electronic Journals
• optical materials
• optical communication system radio
• telecommunications equipment
• telecommunications
• satellite
• telecommunication engineering
• digital communication
• relativistic electrons
• tunable radiation
• fundamentals and practice
• electron lasers
• Solutions manual
• Systems Engineering
• practical systems
• mathematical foundation
• digital communication system design
• analog communications
• Journal of Automation and Control Engineering
• The design of humanoid robot
• C# interface on bluetooth communication
• Humanoid robot via Bluetooth
• Remote control system
• Nuclear engineering and technology
• Development of field programmable gate arrayebased encryption module
• Mitigate man in the middle attack
• Nuclear power plant data communication network
• Field programmable gate array
• Electronic Engineering
• PID controller
• Minimum delay congestion control
• Service communication networks
• Physical network resource limitations
• Atmosphere laser communication channel
• Neural network group
• Optimal weight initialization technology
• Tibetan Plateau
• electrical
• electronic and electrical industry
• electrical engineering
• electrical professional
• specialized dictionaries li arts
• Electrical Engineering Dictionar
• Technologies
• robotic
• mechanical
• anatomy
• Public perception
• hydrogen technologies
• Hydrogen safety communication
• emerging profession
• hydrogen safety engineering
• Knowledge gaps
• future progress
• Hydrogen properties
• hazards