Skip to main content

Fake media - Can blockchain help ?


Fake media - A potential solution using Blockchain ?


Media is an important pillar of democracy but this pillar is loosing trust and getting weaker due to TRP focused approach. This is a big risk for the democratic fabric of  any country. News channel /media industry needs a refinement to remain responsible and trustworthy. 

What is the problem ?

- Media is loosing trust due to fake or misrepresented facts.
- There is no effective regulating body for media.
- Unreliable information going viral online (aka fake news) is one of the biggest threats faced by society.

What is the solution ?

- Need an effective regulating body to control fake news. There is no central agency which can be trusted today.  
- Need a central agency which is unbiased , run automatically and transparently without any human interaction.

How Blockchain can help ?


Step 1:  Create an open community controlled by blockchain 


  • The community should be open for everyone 
  • Each member must be securely validated through KYC.
  • Community members can provide rating between -5 to 5 to any news published by any media house. (Value of community member rating will be based on their credit score. Refer step 2 for details.)
  • Final news rating will be average of all the rating received.
  • Media House will be rated based on the overall aggregated rating received for all the news.
        Say a media house M rated for 5 different news as r1, r2, r3, r4, r5.
        
         Overall Rating of Media House R = (r1 +r2 + r3+ r4+ r5)/5;  

        *News with no rating will not be considered in average calculation.
        *News will all positive rating will also not considered in the average calculation (this is to avoid false positive rating for news with plain facts like India win cricket match).
  • Based on the aggregated rating, Media house will be given Card as  below



 RED     : Rating < 0
AMBER :  0 < Rating <=3
GREEN :  Rating > 3


Step 2 :  Self Regulate Community member
There is a need to identify and isolate paid member or members with personal motives. 
  • Community member will be given a score between  -1 to 10.
  • Net value of the rating (effective rating) given by a community member will be determined by its score. Say Rating is r  and score is s, in that case net value of rating will be r*s.                   So if rating given is 5 but the score of the member is 0.5, in that case effective rating value will be 5*0.5 = 2.5
  • A new member by default will get a score of 0. This means rating given by new member to any news will have no impact on the overall rating of the news (score * rating = 0* rating)
  • Community member will have to earn score to be able to make any real impact, member can earn score by being honest and doing enough due diligence before putting their rating for any news.
How score of community member  will be calculated?
  • Community member can give rating to any news, they can upvote, downvote or remain neutral (no vote or 0)
  • This will work on the assumption that more people supporting any thing is considered correct. (as people will have a risk of loosing score if they try to manipulate)
  • A ratio of upvote vs downvote will be calculated. This ratio will be used to generate score of participating members.
  • Member on the winning side (max supporting votes) will get + score and the loosing side will get -ve score.
         For example for a new 10 upvote and 5 downvote is given to any news.
         score from this news for members who upvoted = (5/10) = 0.5
                                       for member who down voted = (5/5) = -1     
  • Member final score will be cumulative average of each score he/she has got.

Step 3:  Final Rating by considering Community Member score 
Final rating will take care of member score , a member with higher score has high weightage   for his vote compared to member with lower score.

 lets say we have 3 members with score s1, s2 & s3.
they have given 3 ratings r1, r,2 & r3 respectively to a new N.

 overall rating of the news N   = (r1*s1 + r2*s2 + r3*s3) / 3 ;

 And finally media house rating will be calculated as shown in step 1.

Comments

Post a Comment

Popular posts from this blog

What is Proof Of Work in Blockchain?

Proof of Work This is in continuation  of  what is blockchain  .... Now we understand miner needs to solve a puzzle before they get to add new block in the network. This puzzle is called Proof of Work . What is Proof of Work? Yes you have guessed it right, it’s like a rolling a dice. If we (ultra) simplify it, proof of work is similar to rolling a dice. Here it goes. Let’s say only you and I are the only 2 miners in the network. To add a new block, we need a get a number say 5 by rolling a dice. Now let’s say I have 1 dice and you have 1 dice. I roll the dice got 3 (No luck ☹ ) You roll the dice, you got 1 (no luck). I tried again, I got 6, again no luck. You tried again and you got 5! .  Wow lady luck smiles upon you and you get to put the block in the network and you got rewarded for the same. Ok… I felt, I need more processing power, I bought one more dice. So let’s solve the puzzle again. Now I have 2 dice and you have g...
Post a Comment
Read more

PostgresSQL to Amazon RedShift

As in the today's world, volume of data is increasing significantly along with the need of real time analytics on the data. This is driving unprecedented demand for Big Data/No SQL type of databases.  More and more requirements are coming to migrate from traditional (relation) databases to more Morden database. The good news is there are plenty of options like Mongo, Cassandra , Dynamo DB, Redshift , Hadoop etc .  But the challenge is,  they all solve a special needs and selecting right database depends on the actual use cases , type of data etc. Here I am covering a generic use case of migration from Postgres (or any other relational databases) to Redshift. We will focus on 2 key area .. 1) Fundamental Architecture difference between Postgres and Amazon Redshift, especially the way Storage and Scaling is managed. Storage played a very important role when it comes to performance as most of the time its I/O that's impact performances. 2) Best P...
Post a Comment
Read more