Counting function points without humans counting: a dissertation reads HTTP requests to automate SFP
Diego Emanuel Ferreira da Rocha presented at UnB’s Applied Computing graduate program the qualification exam of a dissertation that automates elementary-process identification through HTTP request analysis — reducing the friction of Simple Function Points.
Brasília, October 23, 2025
Simple Function Points (SFP) is today one of the most widely accepted functional size measures in the market — and one of the hardest to apply. Counting is manual, requires comprehensive system documentation, and depends on experts fluent in the method. It is against that friction that Diego Emanuel Ferreira da Rocha’s master’s research is organized, presented in October 2025 at UnB’s PPCA under the advising of Prof. Dr. André Luiz Peron Martins Lanna .
The work proposes automating the identification of elementary processes — SFP’s basic unit — through HTTP-request analysis of running web applications. The idea is elegant: requests reveal user journeys; grouping requests that co-occur reveals functionalities; classifying those groupings by SFP rules yields functional measurement. The method is technology-agnostic — it does not depend on frameworks or languages — and leverages Longest Common Subsequence to handle real variations in observed journeys.
Reported results reach an average variation of 17% against manual counting, close to the 15% acceptable threshold for human-executed measurement. It is a concrete contribution to requirements engineering and applied machine learning : a path to remove from the flow what usually paralyzes SFP adoption.
The qualification took place on October 23, 2025 in Brasília. The examining committee was chaired by the advisor, Prof. Dr. André Luiz Peron Martins Lanna (FCTE/UnB), and composed of Prof. Dr. Monalessa Perini Barcellos (PPGI/UFES) and Prof. Dr. Sérgio Antônio Andrade de Freitas (FCTE/UnB). Prof. Dr. Edna Dias Canedo, PPCA program coordinator, also attended the session.
The full text — written in Portuguese — is available at the UnB Institutional Repository.
About CEDIS: The Center for Studies, Development, and Innovation in Software (CEDIS), linked to the University of Brasília, researches and develops innovative software solutions.