Dominant Meaning Technique for dedicating Social Criminal Twitter Posts

social media like Facebook, Twitter, and LinkedIn has gotten to be a portion of our lives. Cybercrime has ended up an imperative issue, particularly in creating nations. The spread of data with no hazard of being identified and brought leads to an increment in cybercriminals. In the meantime, the huge of information constantly generated from Twitter has made the method of detecting cybercriminals a troublesome task. This paper analyzes how content size such as tweets on Twitter, posts on Facebook, etc. on other social media played in the predict cybercrime. So, in this paper, we try to answer, “What are the fit content sizes that have more effects on accuracy?”. This paper presents a solution based on two techniques: Dominant Meaning DM, and Term Frequency Inverse Document Frequency TF-IDF. This solution constructs super comparable vectors for both pockets negative and positive from different contents that have the same size. These vector plays a vital role to predict pocket for input tweets. To overcome this challenge, we compared the performance of the two mentioned methods. Our results introduced recommendations sizes of content that answered the question of research. However, the recommendation sizes may be disturbed by changes in the technique that generate super comparable vectors. The range of improvement which comes from dominant meaning for precision, recall, and F1 values is 75%, 75%, 70.07% respectively.