(function(doc, html, url) { var widget = doc.createElement("div"); widget.innerHTML = html; var script = doc.currentScript; // e = a.currentScript; if (!script) { var scripts = doc.scripts; for (var i = 0; i < scripts.length; ++i) { script = scripts[i]; if (script.src && script.src.indexOf(url) != -1) break; } } script.parentElement.replaceChild(widget, script); }(document, '

When digital text tricks us: why emojis and look-alike letters matter in language data

What is it about?

This paper shows how emojis, symbols, and look-alike letters can confuse corpus tools, changing word counts and research results. It explains why this happens and how to prepare digital texts so that software analyses what people actually wrote, rather than a distorted version of the data.

Why is it important?

Digital communication now depends on emojis, stylised characters, platform-specific text, and AI-readable data. This paper is timely because these small details can affect not only corpus linguistics, but also digital discourse research, social media analysis, and AI systems trained or tested on online language. It shows that reliable research starts with faithful data: if software misreads the text, the results may look scientific while resting on unstable evidence.

Read more on Kudos…
The following have contributed to this page:
Matteo Di Cristofaro
' ,"url"));