Information Technology Reference
In-Depth Information
“It sounds we are going to employ the Twitter API,” Carl mumbled.
“Yes, Carl, it seems that you are very familiar with Twitter due to daily online
activities.”
“So familiar!” he sighed, rolling her eyes.
“I remember Carl, you had told me that you wanted to work on different projects,”
she admitted. “But to be honest, I think you are one of the most competent interns
who could achieve this task on time owing to your experiences,” she asserted.
“Here I am,” Carl bragged, smiling.
“Then, please stop to whine and keep on listening,” she said, smiling. “The second
task is to classify the tweets that you collected in the first task as positive or negative.”
“Hmmmm, the second task is bipolar sentiment analysis,” he said.
“Yes, it is,” she told. “After sentiment analysis, for each city, your system should
detect the company that has the highest percentage of positive tweets as the third
task.”
“It sounds as if it is a kind of competition,” he said. “We are going to determine
the winner company of each city according to ratio of positive tweets received.”
“Kind of a competition,” she agreed. “We need to pre-”
“How will we
” Carl interrupted. “How will we present the results?” he asked,
“By means of a map or illustration
...
” he added.
“If you didn't interrupt me, I was about to say,” Sandra said, smiling.
“Oops, sorry
...
” he said, looking up.
“As a last step, the winner brands of the day can be monitored on a web page
using the Google Maps”, she told.
“Let me conclude,” he said. “The first step is collecting the tweets, the second one
is applying bipolar sentiment analysis and the last step is developing an interface so
as to present the results,” he muttered.
“Good brief!” she told. “That's what we are going to do.”
“Then, what is the main ambition of this project for our company?” he wondered.
“To detect any bad trend,” she replied.
“What do you mean by detecting any bad trend?” he asked.
“I mean, by means of this system, our company can intervene in any bad trend,”
she told, fingering her pendant.
“Would you please give me an example?” he asked.
“For example,” she answered, “reactions of users to a newproduct could be tracked
and analyzed automatically in order to learn whether users liked or disliked it.”
“That would be really nice for companies,” he said, smiling.
“Definitely,” she agreed. “Time is of the essence, I wish you luck Carl.”
...
2.1 Introduction
As described in Carl's story, sentiments and opinions of customers might be very
important for companies in our times. Social and micro-blogging platforms are
mostly utilized to get this kind of information. Especially, Twitter, whose popularity
Search WWH ::




Custom Search