Search Engine Optimization using Keywords Topic Modeling

When you want to find a certain niche for your idea or product, you need to study the market. Knowing the keywords that your readers or customers use when searching for what they need is a great advantage.

Text network analysis topic modeling as applied to SEO can give you additional insights: how those keywords are connected, what are the most influential ones, and what are the differences between the supply and the demand.

The easiest way to approach the problem is to think of the Google results for a certain search query as the supply and what people are actually searching for as the demand. We can then build two text network graphs: one for the search results, the other one for the keywords used in search queries.

When we put these two graphs together, we can quickly see the keywords that people search for (demand) but do not find (supply). We can also see the relations between them, which gives us a better idea of the main topics we should address when we create a new text or market a new product.

Here is how you can do all this using the SEO text network topic modeling tool in InfraNodus:

Keyword Strategy for SEO using Text Network Topic Modeling

In order to try the search engine optimization app on your data, log in InfraNodus and choose the "SEO" app on the apps screen.

First, use the SEO app to add the keyword query that you are interested to explore. It can be your research topic or a general keyword phrase you want to explore the context for. For example, "topic modeling".

Create a new SEO graph

You will see two graphs superimposed: one that contains the search results (the supply), the other one that contains the related search queries (the demand). Using the Analytics panel you can see what are the main topics related to that search query.

Network graph topic modeling of the search results and queries

You can select the most relevant keywords and topics on the graph and save them into your keywords list.

Add the most relevant keywords for your SEO strategy

You can add more related search keywords using the "+queries" field or more search results for a specific query using the "+results" field.

Add more results into the graph

The most interesting feature is the Missing Keywords: select this option to see what people search for (demand) but do not find (supply). In this case we see that using neural networks for topic modeling is demanded by users but is not represented in search results. So that could be a good niche to explore.

Find the unoccupied niche for your target keywords SEO strategy.

The short video tutorial below explains the workflow for performing search engine optimization using the SEO App of the network visualization tool InfraNodus.

If you are interested in the technical details and would rather do everything yourself, instead of using the app, you can watch this tutorial below:


How Does It Work?

The basic approach is based on our paper: InfraNodus: Generating Insight Using Network Analysis (Paranyushkin, 2019).

We build four graphs:

  1. one graph is the graph of the search results for your keywords query (supply)
  2. the other graph is the graph of the related search queries to your keywords (demand)
  3. the third graph is the graph of your saved keywords
  4. the graph of differences between what people find and what they actually search for

Next, we apply several node ranking and community detection measures, which range the nodes by betweenness centrality (discoursive influence) and groups them by communities (the nodes that are more densely connected together belong to the same topic, have the same color, and are located closer on the graph.

You can then use the aggregate graph to better understand the context around a certain keyword, or to use the Missing Keywords feature to see what people search for but do not yet find — thus identifying a potentially unoccupied niche for your content creation or AdWords strategy.

Try It Yourself

You can try this approach yourself using InfraNodus for any search query you're interested in:

Log In    Sign Up


Custom SEO Optimization Using Network Graphs

There are, of course, many nuances and different sources of data that can be used to improve the SEO strategy.

Contact us if you are interested in a particular use case or if you have any questions about creating a mind map based on your own source data.

Contact Us