Cast your vote
You can rate an item by clicking the amount of stars they wish to award to this item.
When enough users have cast their vote on this item, the average rating will also be shown.
Your vote was cast
Thank you for your feedback
Thank you for your feedback
Chen-Fu Chiang; Reviewer
Saumendra, Sengupta; Advisor
Andriamanalimanana, Bruno; Reviewer
Google Map API
Twitter Live API
Natural Language Processing API
MetadataShow full item record
AbstractThis project basically aims to build a system for the real-time analysis of the trends and public views around the whole world by storing and analyzing the stream of tweets from the Twitter live API which produces a huge amount of data . The tweets, tweet ID, time and other relevant elements are stored into a database and are represented in a map that is being updated in near real time with the help of Google map API. This project also aims to achieve the sentimental analysis of the tweets by sending the tweets to the natural language processing API which in turn processes the tweets using the natural language processing and gives a result If those tweets are positive, negative or neutral in nature. The map clusters tweet as to show where people are tweeting most from according to the sample tweets we get from the streaming API. These clusters will be shown in different colors according to the sentimental evaluation we receive from the sentiment API by Vivek Narayanan which works by examining individual words and short sequences of words (n-grams) and comparing them with a probability model. The probability model is built on a pre labeled test set of IMDb movie reviews. It can also detect negations in phrases, i.e., the phrase "not bad" will be classified as positive despite having two individual words with a negative sentiment. The web service uses a co routine server based on event, so that the trained database can be loaded into shared memory for all requests, which makes it quite scalable and fast. The API is specified here, it supports batch calls so that network latency isn't the main bottleneck. For Instance, if a tweet is negative in evaluation then it is shown in a red color marker on the map, green for positive and grey for the neutral. This analytic will also demonstrate the heat map for all the tweets that are stored in the database which gives a satisfying answer demonstrating from which part of the world are most of the tweets from. In this project we create a dynamic web application with the target runtime environment as Apache Tomcat Server. The server will also be initialized with the context listener which starts running the code to get the tweets into the database till the server is stopped. The most popular trends among worldwide and citywide would be provided in a drop down to be selected from which gives a clear perspective on how each trend behaves. It also offers the public, the media, politicians and scholars a new and timely perspective on the dynamics of the world wide trends and public opinion.