Right Relevance Influencers’ Topical Scores & Rankings

Right Relevance (RR) provides curated information and intelligence on over 50 thousand topics. This includes:

  • Topic relationships including related topics & semantic information like synonyms, acronyms.
  • Topical influencers (over 2.5M) with score and rank.
  • Topical content and information in the form of articles, videos and conversations.

Often asked questions are, what does score for a topic tag assigned to an influencer signify and how is it calculated. For e.g. the scores shown for ‘climate change‘ for the top 4 influencers, along with scores for the additional topic tags, in the image below.

Right Relevance ‘climate change’ influencers

The Right Relevance score of an influencer for a TOPIC represents the authority within the social community for that topic, say for e.g. ‘machine learning’, of that influencer. It is a normalized score ranging from 10 to 100. This numeric influence is then inductively applied to the topical content curated by that individual for measuring relevance.

The process is fully algorithmic and leverages ML, semantic analysis and NLP on unstructured data at scale. It is primarily graph based and involves performing a 2-level proprietary people rank (custom page rank for social graphs):

Stage 1. Global PR to reduce a ~300M nodes graph to ~6M (for now) globally ranked influencers. This is a first level reduction and we don’t expose the scores. It doesn’t have any topical context.

Stage 2. Graph partitioning of the ~6M connected nodes from stage 1 across our ~50K structured topic space using unstructured data assigned to each node. This leads to ~50K per topic sub-graphs, where a secondary PR is applied to determine the topic score for each node in each topical sub-graph. This secondary PR score is normalized to calculate the Right Relevance topic score and rank influencers for every structured topic in our platform. 

Our custom PR algorithm is derived from google pagerank but is specialized for social graphs (instead of links/webpages) with many important differences applicable to social networks.

This measure of influence per topic is termed as ‘topical influence’ or Tribes. More details on our related Insights service including Tribes and Flocks is available here.

Points to note:

  • We dampen followers count, tweet count etc. noisy signals and lay much more focus on the topical network itself.
  • Each influencer can be part of multiple topical sub-graphs aka communities and have a different score, and rank, within each. This is exposed in our apps via scored tags.
  • Other, non structured, topics work via free-form search but the relevance may not be of the same quality. This can be seen by the score ’10’, which, probably poorly done, means we didn’t find a community for the topic.
  • Both our topics and graph are mined and built algorithmically at scale and increasing in number with higher quality with every iteration.

Leave a Reply