Algorithm 1: Match social media posting with sosa:observable property

procedure GETSOSAOBSERVABLEPROPERTY(socialMediaTokens) smTokens ←nlp(socialMediaTokens) sosaObservableProperties ←temperature,humidity,wind,gust,pressure observablePropertiesTokens ← nlp(sosaObservableProperties) tokenSimilarity ← [] dictTokenSimilarity ← [] for eachToken in observablePropertiesTokens do  for eachsmToken in smTokens do  if eachToken.text != eachsmToken.text and eachtoken.similarity(eachsmToken) > 0.50 then   tokenSimilarity.append(eachToken.similarity(eachsmToken))   dictTokenSimilarity[eachToken.text]=eachToken.similarity(eachsmToken)  end  end  if max(dictTokenSimilarity.items(), key ←lambda x : x[1]) is None then  NSOP , 0 ; /* NSOP ← NoSimilarObservableProperty ∗ /  tokenSimilarity.append(eachToken.similarity(eachsmToken));  dictTokenSimilarity[eachToken.text]=eachToken.similarity(eachsmToken) else  mostSimilarObservableProperty,mostSimilarObservablePropertyValue  =max(dictTokenSimilarity.items(),key ←lambda x:x[1]) ; /* return  the most similar observable property(token) with max similarity of all  tokens */  return mostSimilarObservableProperty, mostSimilarObservablePropertyValue  end end end procedure