Feat: add number of keywords without high risk of false positive

[polomarcus]

Goal

In order to have a simpler et faster calculation on link between keyword themes (causes and consequences for example), we have to add 8 integer columns such as number_of_climat_consequence but without high risk of false positive

high risk of false positive

A high risk of false positive keyword such as, eau, become a true positive when it's around a strong keyword such as "climate change".
The problem with that is when we want to observ the link between 2 keyword of different theme the high risk of false postive logic falses our calculation by giving us a too high corrolation.

Columns list ot add

Based on these columns, we have to add `number_of_XXX_no_hrfp" :

• "number_of_changement_climatique_consequences"
• "number_of_changement_climatique_causes_directes"

• "number_of_changement_climatique_constat"
  

• number_of_attenuation_climatique_solutions_directes
• number_of_adaptation_climatique_solutions_directes
• number_of_biodiversite_consequences
• "number_of_biodiversite_causes_directes" +
• "number_of_biodiversite_concepts_generaux"
• number_of_biodiversite_solutions_directes

Json data

Add a boolean inside the keywords_with_timestamp to indicate if the keyword was transformed or not

hrfp: True

After

Once added, we need to reapply these counters from the beggining