Introduction

We are excited to announce a significant update to our Source-Based factor(FLBNess) calculation, an integral part of the Media Impact Value (MIV). 

The MIV is a composite metric, made up of various components, and FLBNess stands out as a crucial element. Specifically, FLBNess measures the extent to which a source is focused on FLB (Fashion, Lifestyle, and Beauty), providing a clear indicator of content relevance in these sectors. 

To calculate the FLBNess of a media source, we follow a two-step process:

Article Count: We count the number of articles focusing on Fashion, Beauty, Jewelry, Watches, and Furniture.
Ratio Calculation: This count is then divided by the total number of articles published by the source. We use the average of this ratio over the last three months to ensure up-to-date relevance, smooth out any anomalies, and provide a more accurate, consistent measure of FLBNess
 

This method provides a clear, current snapshot of the media source's emphasis on these key sectors.

How the Model Works

Data Collection: The model begins by collecting a large volume of articles from various media sources. This includes content from diverse sectors, with a focus on areas like Fashion, Beauty, Jewelry, Watches, Furniture and Other (general content)
 

Deep learning models training: Using transformers and attention mechanisms, the model learns to classify each article. Deep learning-based architectures enable the model to understand article context and categorize content like human reading comprehension.
 

Classification and Tagging: Each new article is then classified based on its content. The model tags articles as relevant to specific sectors (e.g., Fashion, Beauty) or as general content. This classification is crucial for the next step of ratio calculation.

What's Changing?

Advanced AI Integration
 

FLBNess Calculation Enhancement: We've upgraded our FLBNess calculation by incorporating predictions from our newly developed AI model. This model is designed to more accurately identify FLB-focused content, making the FLBNess score even more reliable.

This means you'll get more accurate and relevant MIV for your online coverage!

Impact on you

Thanks to advancements in AI, we are now capable of recognizing FLB content with greater precision, regardless of the language of the article. 

This improvement allows for a more accurate calculation of the MIV, ensuring that sector-specific media are appropriately recognized and rewarded. At the same time, it adjusts the impact of media sources whose audience may not align with your brand. 

Implementation Timeline  🗓  The new FLBNess calculation will be in effect from January onwards. 

Conclusion

This update reflects our commitment to delivering the most accurate, up-to-date information to help your business thrive. 

We're confident that the enhanced FLBNess calculation will bring a new level of precision and relevance to your strategic decisions.