DOES AYLIEN STILL WORK ON RAPIDMINER STUDIO UPDATE
So we recommend to Apple to work on the improvement of this tool and to update the firmware of the Iphone.īeyond this challenge of "the best of the best", I hope that this process will inspire/help RapidMiner's users which have similar We can interpret that the Iphone’s users are not totally satisfied of the auto-correction tool of the Apple’s smartphone. However we saw that in the most frequent words in the negative tweets are « word », « auto », « correctly », « corrects ».
We can see that the words « Apple » and « Iphone » are in the 5 top words of the neutral and positive exprimed tweets.Īfter performing the sentiment analysis and the study of the 5 top words of each sentiment, we can conclude that the Tweeter members have an « overall positive » sentiment of the brand "Apple" and its flagship product the « Iphone ». Connections -> Legacy Connections -> Manage Connections (Legacy) YY. We still have access to the legacy connections and manage them thru. You dont need to convert the legacy connections of Aylien/Rosette into the new connections. We sorted and filtered the 5 top words for each sentiment polarity value : Right now we havent finish re-do all the connections. If we add the neutral and positive tweets, we can consider that 90 % of the tweets express an « overall positive » opinion of the Iphone. 8 tweets are considered as « positive ».28 tweets are considered as « neutral ».4 tweets are considered as « negative ».Sort and filtering of the 5 top words for each sentiment polarity value (negative, neutral, positive) using Sort and Filter Example Range operators.Aggregation of the words according to their sentiment polarity using the Aggregate operator inside a Loop Attribute subprocess.In this step, tweets are tokenized, stop-words are filtered, the words are filtered by lenght (we only retain words whose number of characters is greater than 4). Creation of a words matrix using the Process Documents from Data.This operator determine the sentiment polarity of each tweet according to its content (negative, neutral, positive). Analysis of the sentiment of each tweet using the Analyze Sentiment operator of the AYLIEN extension.Recovery of tweets including the keyword « Iphone » using the Search Twitter operator.add your AYLIEN API keys, if you havent got yours yet, you can sign up here. The RapidMiner’s process execute the following actions : To show how easy it is to use the API with R, were going to run a simple. The goal of this study is to perform a sentiment analysis of 40 tweets including the keyword « Iphone ».įor this study, we will use « RapidMiner » 9.0 (Beta) and its extension « AYLIEN Text Analysis » 0.20. Let me explain this process and the associated results : I allow my self to inaugurate this new "ideas section" with a fictive and experimental sentiment analysis process : There are nuggets in our community and it is an excellent idea to share them !
0 kommentar(er)
