What is Semantic Map
The Semantic Map is a novel way to visually represent knowledge, capturing relationships between documents, keywords, and important topics.
The Semantic Map can analyze a collection of documents. Every dot on a semantic map represents a piece of text from a document. This passage is characterized (or summarized) by several keywords that are closely related to this passage. Let’s take a small area of the map, consisting of a few hundred dots. The question is: Will some keywords and specific documents be especially common in this area? Indeed. So, if two text fragments are close in meaning, their dots will be placed close on the map.
This is the main “wow” of our technology – it transforms semantic proximity to that positional.
Semantically close dots will therefore be grouped in a single area. You can explore the meaning of each area on the map, using keywords and document titles as hints.
The beauty of this map is that it helps to explore a domain of knowledge very fast. In other words, a person who has never worked with a subject area can get to know it by exploring what is discussed in it, which topics it comprises, and how they are connected. Domain experts can also benefit from a better understanding of the domain, for example, find unexpected relationships between topics.
Below, we explain the functionality of the Semantic Map based on English Wikipedia. Semantic maps for other document collections (or domain areas) will have similar functionality, but topics or keywords will be different.
Exploratory Search with the Semantic Map
The semantic map of the domain is shown in the center of the screen. When you view the map, heterogeneity seems evident: in some areas the density of the dots is higher. Such areas are the topics discussed in the documents. The denser the area, the more often the topic is discussed in the documents and the likelier it matters. To make it easier to navigate the map, we have highlighted the most important topics with red or white circles and names. Red circles mark larger topics, while white circles are used for less important topics.
For a more detailed view, zoom in with the mouse scroll. You can also move the map: left-click on it and drag it in the desired direction. When you want to view the required area, click on it. A red selection box will appear. This box will help you to explore what knowledge the selected map area is associated with.
On the right is the Area Explorer, which will help you understand to which area of knowledge the selected area of the map is related. At the top, in the top keywords box, the keywords are listed. Many dots — fragments of the text — fall into the area of the map you are examining. Each text fragment is characterized by several words that are closely related to this fragment. For all dots, the occurrence of associated words is counted. The most frequent ones are shown in the top keywords box.
Below, in the documents box, the names of the documents associated with the selected area of the map are listed. Each dot — a fragment of the text — has been taken from some document. The occurrence of documents is calculated for all selected dots. The most common ones are shown in the documents box. You can click on the name of the document in the list. This document will open in a new browser tab.
Click on the word in the top keywords box. All dots that are close in meaning to the selected word will be highlighted on the map. You can select several keywords at the same time. This way you can find relationships between topics and understand the global semantic structure of the map.
On the left is the Map Explorer, the guide to the map. At the top left is the All Keywords box. It lists the keywords that are most common in the domain area. Click on a keyword to see all dots on the map that are associated with it.
The All Topics box is the list of all topics marked on the map. Click on a topic name to open it on the map.
The Semantic Map is a powerful tool for analyzing collections of texts: it provides text classification, reveals the most important topics discussed in texts, helps to identify relationships between documents, and assists in discovering the most important keywords for topics. It can be created for any collection of documents in any language. The Semantic Map is the cornerstone of the Silk Data Semantic Framework.