How to mine & monitor news data - Analysis of the Metaverse

It’s not very common for an organization to change its name, which is why when Facebook changed its company name to “Meta,” it became one of the most trending topics of discussion. Many news articles got published about the "Why?" behind it and public opinion. The reason for this was so the company could branch out and not be linked to one product. And to be seen as a Metaverse company. This is why Meta was a better name as it represented the organization’s direction. Therefore, in this blog post, we'll look at what's reporting around the Metaverse.

News articles are one of the best ways to follow a trending topic and get more information. However, understanding an overall idea regarding a topic is not quick because it requires reading and analyzing hundreds, if not thousands, of news articles. Also, generating detailed insights based on the news articles is a challenging process. Moreover, this is only achievable with the help of complex analytical tools that require programming expertise.

Dcipher Analytics offers robust text analytics solutions to analyze text data and create insights without any coding knowledge. Furthermore, with its predefined project templates, Dcipher makes it easy to run certain types of analyses. These templates include brand analysis on social media, survey analysis, and multiple document analysis. Moreover, you can create your customized project templates by saving your project pipeline and re-using it with different datasets. This drastically reduces the amount of time spent and also makes text analysis easy.

The ability to import and analyze news data with the pre-defined templates is also possible with the latest updates in Dcipher. Downloading news data is possible without the need for an extra scraping tool. You can download news data with your custom search criteria; click here to learn more. A detailed news media scanning project template that furnishes your workspace by custom preprocessing and enrichment operations and the most relevant workbenches is also available. To learn more about using the news media scanning project template, keep reading. We've used an example of use case analysis about the Metaverse.

1. Create a new project

Firstly, we simply click "Create a new project" to start the process once we get into the platform. Then six options are shown that offer some helpful structured paths. We will then select the "News media scanning" project template.

2. Name the project

We then defined the name of the project as "Metaverse.” Additional project descriptions can also be added.

3. Specify the search criteria

Then we need to fill in the search criteria regarding the news data we want to analyze. We type “metaverse” as a keyword for scanning since we are analyzing the Metaverse. Then, we leave the language as English and the maximum number of articles as 1000 for our criteria. Lastly, we finalize it by naming both the project and the dataset and clicking on “Start.” You can watch the steps in the video below.

4. Read the template details

You will now see a pop-up window that gives detailed information about the template. It contains the pipeline operations & workbenches, and the methodology is shown below:

Operations:

• Title, summary, and article text are merged into a single column

• Duplicate and almost duplicates articles are removed

• URLs, HTML tags, hashtags, and handles are removed from the article content if existing

• Articles are split into smaller segments via smart segmentation operation

• Segments are tokenized into words and phrases with lemmatization, stop word removal, part-of-speech tagging, and Named Entity Recognition options enabled

• Topic modeling is run and each segment is assigned to one or more topics

Workbenches:

• Bar Chart: This workbench displays the top 20 news sites based on the number of published articles in your dataset

• Text View: This workbench displays news articles with their title, summary, publisher source, and published date.

• Foam Chart: This workbench generates a foam tree where topics are displayed on the surface level and relevant words are displayed inside each foam. The sizes of the foams represent the strength of the topic in your dataset and the relevance of a word to the topics respectively.

• Bump Chart: This workbench displays how the rank of each topic detected on segments changes over time at daily granularity.

• Scatter Plot: This workbench maps segments extracted from news articles to a 2-dimensional plane based on semantic similarities between the segments and generates a landscape of documents. Similar texts appear close to each other. Additionally, data points in the landscape are colored based on the most dominant topic they match.

• Bubble View: This workbench generates a network of words, entities, and phrases extracted from the news articles based on contextual similarities between them. This network provides insights into how tokens are connected and related to each other.

We highly suggest you read this carefully to understand the template and the analysis output better. You can reach these instructions any time you need by clicking the “?” button at the top-right of the page if you close the descriptive window to continue your analysis.

In the meantime, the analysis continues to run operations and creates determined outputs in each workbench.

5. The final output

The workbenches are populated with output data once the tasks are complete. You can view each one of them and observe the generated insights as a result of the analysis.

Some of the interpretations are as follows for the Metaverse analysis:

• 100+ news data sources were scraped, and the top news sources are Bitcoin Etherium News, Knowledge News, and Daily Advent.

• The most discussed topics based on topic detection and summary operation are about the potential of the Metaverse to create an increasing overlap of our physical and digital lives and the role that Bitcoin and other cryptocurrencies will play in the Metaverse.

• Moreover, another part of the overall discussion is around the underlying technological framework that the Metaverse comprises (like VR and AR) and the resulting digital assets (like NFTs) that are created in this virtual world.

The Metaverse analysis reveals many interesting topics. Let us know the part that you found surprising! And to not miss out on our future suggestions, make sure to follow us on our LinkedIn page as well.

Additionally, If you’d like to learn more about other types of charts used in different examples, you can also check out our previous blog posts:

• Dr. Anthony Fauci’s emails: what can we learn from a deep-dive with Natural Language Understanding?

• Where Do Americans Want to Travel This Summer? The Answer Lies In Our 5 Step Analysis From Social Media!

• Find The Best Gift Ideas for this Father’s Day in 6 Quick & Easy Steps!

• Four ways of visualizing free-form survey responses

Get started!

To access our text analytics toolbox and try out analyzing free-form text responses in Dcipher Analytics, sign up for a free trial.