RAHMA, AISYAH FIRDAUSI (2021) ANALISIS SENTIMEN HASHTAG KULINER DI INDONESIA MENGGUNAKAN NAIVE BAYES. Undergraduate thesis, UPN"VETERAN" JATIM.
|
Text (COVER)
Cover.PDF Download (948kB) | Preview |
|
|
Text (BAB 1)
1.PDF Download (253kB) | Preview |
|
![[img]](http://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (BAB 2)
2.PDF Restricted to Registered users only Download (637kB) |
|
|
Text (BAB 3)
3.PDF Restricted to Registered users only Download (164kB) |
|
|
Text (BAB 4)
4.PDF Restricted to Registered users only Download (604kB) |
|
|
Text (BAB 5)
5.PDF Download (9kB) | Preview |
|
|
Text (DAFTAR PUSTAKA)
Dapus.PDF Download (281kB) | Preview |
Abstract
Along with the development of the times, new technology more and more popping up. The number of new technologies resulting in an increase in the amount of data generated. The emergence of social media called twitter in 2006 also followed contributed to an increase in the amount of existing data. Use Twitter social media can make users share about daily life or information with colleagues or relatives in real time. The very high number of twitter users can become a means for the community to make buying and selling (promotion), it can also be to write what feelings felt, what products are being used and also reference for places to eat or culinary in Indonesia. That matter is one of the reasons why twitter is social media appropriate in conducting sentiment analysis. Sentiment analysis itself needs to be done because of the excessive use of social media increasing so that it can affect the development public opinion. Creating a sentiment analysis model with the culinary hashtag on twitter in Indonesia it is built using python. The research was conducted by carrying out the stages of data collection, preprocessing, data labeling (positive, negative, and neutral), text weighting, data sharing, Naive Bayes Classifier, evaluation classification model and visualization. This model is built using two scenarios. On text weighting stages, the first scenario is to use The next Bag of Words in the second scenario uses TFIDF. The dataset used has a total of 748 tweets taken within January 1, 2020 to November 30, 2020. The dataset has been labeled with 3 labels, namely positive, negative and neutral. Dataset retrieval is taken using keywords that are has a culinary hashtag. The best classification modeling results is to use Naive Bayes Multinomial on scenario 1 with an accuracy value of 0.71, an average precision of 0.80, an average recall of 0.53, and an average f-score of 0.53. Keywords: sentiment analysis, classification, Twitter, Naive Bayes
| Item Type: | Thesis (Undergraduate) | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||
| Subjects: | Q Science > QA Mathematics > QA76.76.E95 Expert Systems | ||||||||
| Divisions: | Faculty of Computer Science > Departemen of Information Systems | ||||||||
| Depositing User: | Mujari Mujari | ||||||||
| Date Deposited: | 22 Jun 2021 03:21 | ||||||||
| Last Modified: | 22 Jun 2021 03:21 | ||||||||
| URI: | http://repository.upnjatim.ac.id/id/eprint/2060 |
Actions (login required)
 |
View Item |