Keyword extraction on Pulsar is the process of automatically identifying and extracting the most common words in a post or article. Similar to Entity or Topic analysis, the aim of keyword extraction is to condense and summarise the main concepts in a search, making it easier to understand what's being discussed. The main difference however, is that keyword analysis is simply based on the frequency of that term.

Keyword extraction can be especially useful for large datasets or collections of text, where manual analysis of each document would be impractical or impossible. By automating the process of identifying common keywords, it can help researchers, businesses, and brands to quickly and efficiently analyse and understand large volumes of text data.

Similar to the Hashtags, Entities and Topics on TRAC, you’ll be able to see Keyword analysis in various charts, under Content Insights. Let’s take a look at each one of them in more detail.

The Treemap visualisation provides insights into the channels where certain keywords are used most frequently. The size of the tile represents the prevalence of that keyword being used on a particular channel. It also helps to identify if certain keywords are over-indexed or under-indexed on some data sources. For instance, the screenshot below shows that cocktail is a highly popular keyword across all channels. However, the Treemap highlights that the keyword over-indexes by +5.44% on Pinterest compared to other data sources, which is an interesting insight.

When looking at keywords displayed in a Sentiment Word Cloud, you can understand the most common keywords in a search and the sentiment associated with those terms. The bigger the size of the keyword, plus the more central it is in the graph, then the greater the number of posts and articles containing that keyword.

Similar to the Sentiment Word Cloud, when looking at keywords displayed in an Emotions Word Cloud, you can understand the most common keywords in a search and the emotions associated with those terms. The bigger the size of the keyword, plus the more central it is in the graph, then the greater the number of posts and articles containing that keyword.

When looking at keywords displayed in a Segments or Network chart, you can start to identify groups of terms that tend to be associated or discussed together in the same conversation. This can be useful to help understand how specific keywords are closely related. Related keywords are grouped together into distinct segments, and the clustering algorithm we apply determines the relevance and importance of each keyword within a given segment, helping you uncover the dominant themes in a dataset.

By analysing the keywords presented in a Stream graph, you can keep track of how the conversation surrounding specific themes develops over time. This enables you to recognise when a particular keyword first appeared, as well as when the discussion waned and perhaps resumed. By linking this data with pertinent media events, you can attain valuable insights into the reasons why certain keywords were used in the discourse.

Sometimes known as a chord diagram, the keywords Bundle chart is a graphical representation of the relationships between the different terms in a dataset. It's a great way to visualise the inter-relationships and flows between the keywords as arcs or chords, that connect the terms. Each keyword is represented as a segment around the perimeter of the circle, with the chords connecting the keywords representing the degree of overlap, or connection between them. The bigger the segment around the perimeter of the circle is, then the more connections that keyword has with other terms in the bundle chart.

By using the insights gained from keyword analysis, you can inform content creation and recommendations based on the most commonly used keywords and hashtags within your dataset. This can help you identify trends and patterns, as well as areas for improvement in your content strategy.

​

​
​

​

​