Methods to classify objects into two or more classes are at the core of various disciplines. When a set of objects with their true classes is available, a supervised classifier can be trained and employed to decide if, for example, a new patient has cancer or not. The choice of performance measure is critical in deciding which supervised method to use in any particular classification problem. Different measures can lead to very different choices, so the measure should match the objectives. Many performance measures have been developed, and one of them is the F-measure, the harmonic mean of precision and recall. Originally proposed in information retrieval, the F-measure has gained increasing interest in the context of classification. However, the rationale underlying this measure appears weak, and unlike other measures, it does not have a representational meaning. The use of the harmonic mean also has little theoretical justification. The F-measure also stresses one class, which seems inappropriate for general classification problems. We provide a history of the F-measure and its use in computational disciplines, describe its properties, and discuss criticism about the F-Measure. We conclude with alternatives to the F-measure, and recommendations of how to use it effectively.

More from our Archive

• DOI: 10.3390/computation11070143 2023

  Mathematical Modelling of Tuberculosis Outbreak in an East African Country Incorporating Vaccination and Treatment

  Kayode Oshinubi, Olumuyiwa James Peter, Emmanuel Addai, Enock Mwizerwa, Oluwatosin Babasola, Ifeoma Veronica Nwabufo, Ibrahima Sane, Umar Muhammad Adam, Adejimi Adeniji, Janet O. Agbaje

• DOI: 10.7717/peerj-cs.1405 2023

  Addressing Internet of Things security by enhanced sine cosine metaheuristics tuned hybrid machine learning model and results interpretation based on SHAP approach

  Milos Dobrojevic, Miodrag Zivkovic, Amit Chhabra, Nor Samsiah Sani, Nebojsa Bacanin, Maifuza Mohd Amin

• DOI: 10.1145/3605889 2023

  Arabic ChatGPT Tweets Classification Using RoBERTa and BERT Ensemble Model

  Muhammad Mujahid, Khadija Kanwal, Furqan Rustam, Wajdi Aljedaani, Imran Ashraf

• DOI: 10.3390/computation11070126 2023

  Response Spectrum Analysis of Multi-Story Shear Buildings Using Machine Learning Techniques

  Manolis Georgioudakis, Vagelis Plevris

• DOI: 10.7717/peerj-cs.1483 2023

  MGACA-Net: a novel deep learning based multi-scale guided attention and context aggregation for localization of knee anterior cruciate ligament tears region in MRI images

  Mazhar Javed Awan, Mohd Shafry Mohd Rahim, Naomie Salim, Haitham Nobanee, Ahsen Ali Asif, Muhammad Ozair Attiq

• DOI: 10.3390/computation11070130 2023

  Applications of Modified Bessel Polynomials to Solve a Nonlinear Chaotic Fractional-Order System in the Financial Market: Domain-Splitting Collocation Techniques

  Mohammad Izadi, Hari Mohan Srivastava

• DOI: 10.4218/etrij.2023-0050 2023

  Miniaturized meander‐line dual‐band implantable antenna for biotelemetry applications

  Rajagopal Kangeyan, Madurakavi Karthikeyan

• DOI: 10.3390/computation11070129 2023

  Computation of the Exact Forms of Waves for a Set of Differential Equations Associated with the SEIR Model of Epidemics

  Nikolay K. Vitanov, Zlatinka I. Dimitrova

• DOI: 10.3390/computation11070127 2023

  Analytical and Numerical Results for the Diffusion-Reaction Equation When the Reaction Coefficient Depends on Simultaneously the Space and Time Coordinates

  Ali Habeeb Askar, Ádám Nagy, Imre Ferenc Barna, Endre Kovács

• DOI: 10.3390/computation11070124 2023

  Fast Operation of Determining the Sign of a Number in RNS Using the Akushsky Core Function

  Egor Shiriaev, Nikolay Kucherov, Mikhail Babenko, Anton Nazarov