Word Cloud FAQ
Written by Raul Garreta
Updated over a week ago

How does the Word Cloud work?

  1. Upload a CSV file or paste text directly into the tool.

  2. Click "Generate Cloud" to create your word cloud.

  3. Customize your word cloud. Change theme, font, color, and word quantity.

  4. Download word cloud images, plus frequency and relevance scores.

What is frequency?

Using a relative frequency (frequency normalized by the maximum value) in which valid terms in other terms are taken into account. For example, if you have "text" and "current text", the weight of "text" is removed from the weight of "current text". So the frequency of the word "text" is calculated only when it appears as a single word, and not as a collocation.

If a text is more than 2400 characters long, the TF-IDF is calculated using an English dictionary of IDFs, or a list of unique words and collocations. If the term is not in the dictionary, it is assigned a low IDF (or low relevance) score of 0.3.

The result is a set of keywords in a text, with their weights for that particular text. The term "book" might be weighted differently in different documents.

Can the word cloud make shapes?

It’s not possible to create word clouds with specific shapes but you can change the layout of your word cloud by altering theme and word quantity.

Can we edit the word cloud?

You can customize your word cloud using the toolbar, which sits just above your word cloud results. Choose a theme, change text color, pick a font, change word quantity, and edit your text.

The size of a word is proportional to the number of times it appears in your data so, unless you add the word more often to your text, it won’t appear larger. On the right, check the frequency and relevance scores to see how often words appear.

Why don't more words appear?

MonkeyLearn only shows the most frequent words that appear in your text. If you enter a small amount of text such as a paragraph or a note it will only display the words that commonly occur.

MonkeyLearn’s algorithms calculate how frequently a term occurs in a document divided by the total number of terms in the document. It then multiplies the term frequencies (TF) with Inverse Document Frequencies (IDF), a weight indicating how often a word is used across documents. The more often a word appears, the lower its IDF score.

MonkeyLearn also prevents stopwords, available from a fixed list in English only, from appearing in your word cloud.

Can I make a transparent background?

Yes, you can. On the menu at the top select background color > transparent.

Can I create a "phrase cloud"?

No, MonkeyLearn’s intelligent algorithms automatically detect collocations, words that often go together, so that you can create word clouds with phrases. Examples of collocations are “save time”, “word cloud”, and “artificial intelligence”, and often provide more context than single words. But you can't force the word cloud to group words or phrases.

Is the word cloud available through your API?

The word cloud tool is not available through our API. You can use our keyword extractor tool to extract the most important keywords and phrases from your text using our API.

Is there a maximum amount of text that can be analyzed?

Yes, the word cloud tool is limited to 200000 characters.

Did this answer your question?