Framework Based on Simulation of Real-World Message Streams to Evaluate Classification Solutions

被引:0
作者
Hojas-Mazo, Wenny [1 ]
Macia-Perez, Francisco [2 ]
Martinez, Jose Vicente Berna [2 ]
Moreno-Espino, Mailyn [3 ]
Fonseca, Iren Lorenzo [2 ]
Pavon, Juan [4 ]
机构
[1] Univ Tecnol La Habana, Fac Ingn Informat, Dept Inteligencia Artificial Infraestruct Sistemas, Calle 114 11901,Entre 119 & 127, Marianao 19390, La Habana, Cuba
[2] Univ Alicante, Dept Comp Sci & Technol, Alicante 03690, Spain
[3] Inst Politecn Nacl, Ctr Invest Comp, Ciudad De Mexico 07738, Mexico
[4] Univ Complutense Madrid, Inst Tecnol Conocimiento, Madrid 28040, Spain
关键词
classification; evaluation; non-stationary message streams; simulation;
D O I
10.3390/a17010047
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Analysing message streams in a dynamic environment is challenging. Various methods and metrics are used to evaluate message classification solutions, but often fail to realistically simulate the actual environment. As a result, the evaluation can produce overly optimistic results, rendering current solution evaluations inadequate for real-world environments. This paper proposes a framework based on the simulation of real-world message streams to evaluate classification solutions. The framework consists of four modules: message stream simulation, processing, classification and evaluation. The simulation module uses techniques and queueing theory to replicate a real-world message stream. The processing module refines the input messages for optimal classification. The classification module categorises the generated message stream using existing solutions. The evaluation module evaluates the performance of the classification solutions by measuring accuracy, precision and recall. The framework can model different behaviours from different sources, such as different spammers with different attack strategies, press media or social network sources. Each profile generates a message stream that is combined into the main stream for greater realism. A spam detection case study is developed that demonstrates the implementation of the proposed framework and identifies latency and message body obfuscation as critical classification quality parameters.
引用
收藏
页数:15
相关论文
共 30 条
[1]   Privacy and Security Issues in Online Social Networks [J].
Ali, Shaukat ;
Islam, Naveed ;
Rauf, Azhar ;
Din, Ikram Ud ;
Guizani, Mohsen ;
Rodrigues, Joel J. P. C. .
FUTURE INTERNET, 2018, 10 (12)
[2]   THE IMPACT OF SOCIAL MEDIA PEER COMMUNICATION ON CUSTOMER BEHAVIOUR - EVIDENCE FROM ROMANIA [J].
Anastasiei, Bogdan ;
Dospinescu, Nicoleta ;
Dospinescu, Octavian .
ARGUMENTA OECONOMICA, 2022, 48 (01) :247-264
[3]  
[Anonymous], 1999, Data preparation for data mining
[4]  
[Anonymous], Jubatus: Distributed Online Machine Learning Framework
[5]  
Apache Software Foundation, 2021, Apache Spark-Unified Analytics Engine for Big Data
[6]  
Bifet A, 2010, J MACH LEARN RES, V11, P1601
[7]   Wild patterns: Ten years after the rise of adversarial machine learning [J].
Biggio, Battista ;
Roli, Fabio .
PATTERN RECOGNITION, 2018, 84 :317-331
[8]  
Biggio Battista, 2013, Machine Learning and Knowledge Discovery in Databases, P387, DOI [DOI 10.1007/978-3-642-40994-3, 10.1007/978-3-642-40994-3_25, DOI 10.1007/978-3-642-40994-3_25]
[9]   Aligning adoption messages with audiences? priorities: A mixed-methods study of the diffusion of enterprise architecture among the US state governments [J].
Bui, Quang Neo ;
Lyytinen, Kalle .
INFORMATION AND ORGANIZATION, 2022, 32 (04)
[10]   Looking for the Sustainability Messages of European Universities' Social Media Communication during the COVID-19 Pandemic [J].
Bularca, Maria Cristina ;
Nechita, Florin ;
Sargu, Lilia ;
Motoi, Gabriela ;
Otovescu, Adrian ;
Coman, Claudiu .
SUSTAINABILITY, 2022, 14 (03)