1. You will get a lot of irrelevant data along with the relevant ones and sorting may take a good amount of time.

2. Data may be presented in other forms like pictures or graphs.

3. The structure of data may vary from company to company i.e. companies may use different words to convey the same meaning.

4. If you use the unweighted method, you will get some benefits. But if you use the weighted method, you'll definitely face some problems in scoring.

RapidMiner is in my opinion the best way to approach this type of data. The learning curve is not steep and there is a huge community ready to help you. Additionally lots of YT videos are available.

https://community.rapidminer.com/

Using Automodel is fun: https://rapidminer.com/products/auto-model/

Existing data mining software seems good that you mentioned for the purpose of mining. Most of the organizations are using it as well as personal use for time consuming issues and less effort. But if you want to know the every pre-processing phase and output generated phase of the data mining what is actually happen in backend, i suggest you raw python for data science. Note that, in case of using tools, somtimes i'm facing problem in NLP.

Dear Dr.

Please get a look

https://community.rapidminer.com/

best regards

A desvantagem é o alto custo de licenciamento. Você pode simplificar suas tarefas fazendo uso o Pycharm (python).

Thank you Dr. Oberdan Costa for your response.

Thank you Dr. Hermann Gruenwald for your response.

Thank you Dr. Sarmad Sobhi Al Qassar for your response.

Thank you Dr. F.M. Javed Mehedi Shamrat for your response.

Thank you Mr. Raihan Sobhan for your response. I really appreciate those points.