From source code identifiers to natural language terms

Abstract

a b s t r a c tProgram comprehension techniques often explore program identifiers, to infer knowledge about programs. The relevance of source code identifiers as one relevant source of information about programs is already established in the literature, as well as their direct impact on future comprehension tasks.Most programming languages enforce some constrains on identifiers strings (e.g., white spaces or commas are not allowed). Also, programmers often use word combinations and abbreviations, to devise strings that represent single, or multiple, domain concepts in order to increase programming linguistic efficiency (convey more semantics writing less). These strings do not always use explicit marks to distinguish the terms used (e.g., CamelCase or underscores), so techniques often referred as hard splitting are not enough. This paper introduces Lingua::IdSplitter a dictionary based algorithm for splitting and expanding strings that compose multi-term identifiers. It explores the use of general programming and abbreviations dictionaries, but also a custom dictionary automatically generated from software natural language content, prone to include application domain terms and specific abbreviations. This approach was applied to two software packages, written in C, achieving a f-measure of around 90% for correctly splitting and expanding identifiers. A comparison with current state-of-the-art approaches is also presented.

Searching source code fragments using incremental clustering

“…This is an expected result, since the identifiers are the basic element of each language. There are a lot of other works that build source code analysis methods based only on the source code identifiers . The next four elements are also present in most vectors.…”

Measuring the understandability of WSDL specifications, web service understanding degree approach and system

“…First, the identifier is retrieved by using the method retrieveIdentifier(). Once the identifier is available, the split algorithm is applied; in this case samurai() splitter is the used algorithm [11], [22], [12]. Samurai returns a set of words (that compound the identifier under analysis) which are searched in the domain specific dictionary (findInDictionary()) with the goal to verify if it is a valid word or not.…”

Perbaikan Metode Rekomendasi Diskusi Pemrograman dengan Normalisasi Identifier Menggunakan Lingua::IdSplitter

“…Hard split dilakukan untuk memisahkan identifier berdasarkan karakter tertentu seperti tanda garis bawah (underscore) atau berdasarkan aturan penulisan camel case. Sementara itu, soft split akan memisahkan identifier yang tidak terlalu terlihat (tanda) pemisahnya [9]. Ilustrasi proses pemisahan dan ekspansi ini dapat dilihat pada Gambar 2.…”

“…Mekanisme pemisahan dan ekspansi identifier [9] [9] Otomata pada Gambar 3 digunakan untuk menghitung skor kandidat hasil pemisahan. Dengan melihat skor kandidat kata pada Tabel 1, didapatkan bahwa kandidat term time dan sort adalah term yang paling baik untuk menjadi hasil pemisahan atas identifier timesort.…”

“…Sementara programmer lain lebih suka menulis identifier dalam bentuk lengkap dibandingkan dengan bentuk singkatan. Mengatasi hal tersebut, [9] mengembangkan Lingua::IdSplitter. Algoritma tersebut memiliki kemampuan untuk memisah identifier yang umumnya terdiri atas komposisi term serta mengekspansi term singkatan pada identifier menjadi term lengkap, baik yang menggunakan model penulisan all lowercase maupun camel case.…”