Since there are several problems that have been identified, this data does not have negative and positive comment classification. This could become a problem for two reasons. Firstly, the sophisticated statistical approach cannot be in use to make prediction without this classification, which should have it for determining confidence interval. Secondly, as held in high regard of data integrity, one should not rely on different data to train for statisitcal purpose. That is, using the product-related data to train machine learning for interpreting the specific parts of text in political-related data could result in measurement error.

The created_at variable that should be date datatype does not behave as a such. Since the lobbyists are more likely to be interested in particular legislators who change their poltiical stance over time, this datatype is, thus, useful. The problem is that this variable does not behave like this datatype. Perhaps, this variable is encrypted, which allows authorized users only to access this kind of information. Before the hypotheses explain, the next slide will explain the data approach.

There are two different programming languages for sentimental analysis and data pipelining. The R is powerful and it is perfectly capable of processing natural language in complex sense. Not only that, I have experience working with this language for more than one year. The SQL integrates in Apache Spark, which is a data frame that can speed up the data retrival and reading.

As mentioned earlier, the R excels at natural language processing so that the appropriate lexicon can be applied for identifying the representatives’ and senators’ political stance for climate change. The objective for this assessment is to identify certain congressional members who might exhibit their political ambivalence, which is effective for lobbyists.

The R will develop the lexicon with full of names that this text mining aims to extract only from names of congressional members. Then, the new dataset is in use to count the highest frequency of people they talk to. The goal is to develop the interactive networking of congressional members who often talk to.

The assumption is that legislators who use “climate action” term without contradiction transition attempt to show their stance in clear way. That is, if the congressional members retweet or repeatedly tweet the comments with term “climate action” and without contradiction trasnition at five times, then their stance is firm.

If the congressional members talk to one member, then other members in same party might talk to same one. In other words, the bike wheel and networking may be resemblant. That is, all communications link to each party’s leader.

If the sentimental analysis applies, then Republican party is more likely to have same tone but Democrat party might have mixed up tone.