Abstract
Social networks are becoming more and more real with their power to influence public opinions, election outcomes, or the creation of an artificial surge in demand or supply. The continuous stream of information is valuable, but it comes with a big data problem. The question is how to mine social text at a large scale and execute machine learning algorithms to create predictive models or historical views of previous trends. This paper introduces a cyber dictionary for every user, which contains only words used in tweets - as a case study. Then, it mines all the known and unknown words by their frequency, which provides the analytic capability to run a multi-level classifier.
Original language | English |
---|---|
Title of host publication | International Conference on Electrical, Computer, Communications and Mechatronics Engineering (ICECCME 2021) |
Place of Publication | Piscataway, NJ |
Publisher | IEEE |
Number of pages | 5 |
ISBN (Electronic) | 9781665412629 |
ISBN (Print) | 9781665429436 |
DOIs | |
Publication status | Published - 7 Oct 2021 |
Event | International Conference on Electrical, Computer, Communications and Mechatronics Engineering - hybrid conference, Mauritius Duration: 7 Oct 2021 → 8 Oct 2021 http://www.iceccme.com/ |
Conference
Conference | International Conference on Electrical, Computer, Communications and Mechatronics Engineering |
---|---|
Abbreviated title | ICECCME 2021 |
Country/Territory | Mauritius |
Period | 7/10/21 → 8/10/21 |
Internet address |
Keywords
- Losless mining
- Sentiment analysis
- Social media analytics
- Social network