Visual Google Search: Explore the Context around your Topic of Interest
Suppose you're searching for something that you don't know so much about. Your search queries will tend to be quite general. You might want to use Google to explore a certain topic and the context around it, but its interface is not really made for this task.
That's why we created a visual search system using text network visualization tool InfraNodus. It visualizes the first 40 search results from Google (or Google Scholar) and presents them as a graph. On this graph you can see not only the most prominent keywords, but also — how they are connected.
As a result, you get a quick and direct understanding of the context and the main topics related to your Google search. You can use this data to learn more about a certain field or to create relevant content.
Visual Google Search with a Network Graph in 4 Steps
In order to use visual search feature, log in InfraNodus and choose the Visual Search app on the Apps screen of InfraNodus. Then, follow these steps:
First, make a search query using a general keyword phrase.
You will see a graph where the most prominent keywords are bigger and the keywords that belong to the same topic have the same color and are closer, so you can better understand what context the keywords appear in.
Click on the most interesting keywords and make a new search to add the results for the new keyword combination into the graph and have a better representation of this field or domain of knowledge.
When you want to open the actual search results, choose the most relevant nodes on the graph and InfraNodus will filter the search results that contain all of these terms.
The short video tutorial below explains this process in detail and shows the workflow that you can follow both for research and search engine optimization purposes.
How Does It Work?
The basic approach is based on our paper: InfraNodus: Generating Insight Using Network Analysis (Paranyushkin, 2019).
Step 1: First, we import the Google search results for your search query. Then we extract the lemmas and each unique word is represented as a node while co-occurrence of words (within a sentence) is represented as a graph edge.
Step 2: Next, we apply several node ranking and community detection measures, which range the nodes by betweenness centrality (discoursive influence) and group 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.
Step 3: We also remove your search terms from the graph (you can re-add them from the top right corner), so you see the context behind those terms. Using the Analytics panel (right) you can see the main topics related to your search query and the most relevant keywords.
Step 4: If you decide to explore a certain topic further, select it or click the nodes in the graph and click the "Add to Graph" dialog at the bottom left. You will then have the search results for these new terms added to your old graph.
Try It Yourself
You can try this approach yourself using InfraNodus for any search query you're interested in:
Log In Sign Up
Search through Custom Databases
The example above works with the Google search, however, as part of our Enterprise plan we can implement this same search on any data and text corpus: an organizational knowledge base, scientific paper abstracts, marketing surveys, reviews, customer feedback.
This approach will help you have a better visual understanding of your data and obtain the insights that are not available with conventional tools.
Contact Us