Business Understanding : The topic chosen is fresh graduate. The topic was chosen because currently there are many students who have just graduated/graduated in these months. From this topic, a classification can be made related to articles/tweets about fresh graduates, to find out how many categories can be created and details related to those categories.
Data : 208 Tweets
Text Preprocessing
- Case Folding : remove punctuation and non-letter characters
- Stopword : remove unnecessary conjunctions in sentence analysis
- Slang Word : handle slang words contained in sentences
Text Data Analysis
- most common word : The most frequently occurring words
- kerja : 127
- lowongan : 77
- pt : 39
- Distribusi bi-gram : Frequency distribution of each bigram (2 words) in a sentence
- (lowongan, kerja) : 66
- (sma, smk) : 27
- (kerja, program) : 25
Model : KNN with 4 clusters
-
Wordcloud class 1
-
Wordcloud class 2
-
Wordcloud class 3
-
Wordcloud class 4
