Social Networking Behavior Analysing

Project Abstract / Summary : Existing model:-
We designed and implemented a behavior analyzing system that uses twitter data to analyze behavior in real time. Twitter data is continuously downloaded using Twitter streaming API. Various tweets attributes like username, language of the tweet, month, year and location from where it was posted are extracted along with the tweet messages. These attributes, along with the original tweet message are then stored in a database. Two text files are stored which contains positive and negative words list. By comparing tweets of followers and other peoples ,tweets are analyzed to know behavior about that person. All the output data is visualized as an interactive map, bar charts, and word clouds. Our aim was to analyze the data present on social media. People often post messages about interesting news, their daily activities, thoughts and feelings. This may help to analyze their behavior and behavior of other peoples towards him/her. It makes easy to analyze positive and negative attitude from their tweets Twitter message, when used in combination, helps increase the accuracy of our prediction. Project included the extraction of tweets using Twitter API and then analyzing them in many possible ways. Apart from this, a tool named Cirrus has been used to take the analyzing process to another step. Foranalyzing, Cirrus tool used for analyzing the extracted data, is a word clouddisplaying the frequency of words appearing in a corpus. This tool not only allows usto remove the stop words, but it also helps us to analyze the tweets frequency monthwise to know activeness on twitter is increasing or decreasing .
Here, cirrus was used for two type of analysis:
1.To know relative frequency of positive words used in tweets verses negative words in tweets.
2. To know frequency of tweets month wise, frequency of hash tags in entire document and frequency of keywords which makes able to analyze behavior and social awareness of any person.

Proposed Model:-
The project is to analyze the data present on social media. People often post messages about interesting news, their daily activities, thoughts and feelings. This may help to analyze their behavior and behavior of other peoples towards him/her. It makes easy to analyze positive and negative attitude from their tweets. Celebrities publicity or how famous they are, what people think about them and how positive and negative image they have ,all this can be analyze fromsocial data.
I will explore this project by getting follower’s details of popular faces and to analyze the ratio of followers country and how’s their behavior is varying. I will also work to get geo location of tweets so that we can get all those places which are most frequently visiting and which areas are close to terrorists and belongs to terror areas. It will also help police to know those peoples who belongs to terrorists and who can be in danger. I plan to gather additional online data (e.g., Facebook, blogs data, news feeds) for real-time behavior analyzes. Another possible research direction is to determine and implement machine learning algorithms for analysis of tweets, their pre-processing and further integration of the modules into one complete using Hadoop or R-programming.

Why did you choose to work on this project topic : Social media like Facebook, Twitter is a continuous source of real time information. The data present on social media can be used as knowledgeable information to serve various domains like depicting natural calamity (earthquakes etc.).Human behaviors are multifarious and myriad in nature. It is a challenging task to envisage and learn the human behavior from daily routine activities. Detection and prediction of human behavior from daily life activities is a challenging task. People can have both regular and varying daily life routines that make it a burning topic nowadays in social research circles. Modeling human behavior such as individual routines from proximity data and social relations with gathered data of daily life activity patterns is an emerging realm in Ubiquitous Computing. Applying data mining techniques to social media can yield interesting perspectives about individual human behavior, detecting hot issues and topics, or discovering a group and community. This aims at implementation of different views of data like facebook posts, status etc on Machine Learning ensemble methods using different heterogeneous sets to extract useful information. It is based on short text classification through which certain texts will be marked positive or negative based on the presence of certain behavior related keywords. It relates the unstructured information which is processed by KDD process to generate useful information.

Project Highlights : We designed and implemented a behavior analyzing system that uses twitter data to analyse behavior in real time. Twitter data is continuously downloaded using Twitter streaming API. Then the tweet texts are extracted along with the time stamps, and user locations and stored in a database, which is used for further analysis in the geographical and text terms. All the output data is visualized as an interactive map, bar charts, and word clouds. Our aim was to analyze the data present on social media. People often post messages about interesting news, their daily activities, thoughts and feelings. This may help to analyze their behavior and behavior of other people towards him/her. It makes easy to analyze positive and negative attitude from their tweets Twitter message, when used in combination, helps increase the accuracy of our prediction.This project includes the extraction of tweets using Twitter API and then analysing them in many possible ways. Apart from this, a tool named Cirrus has been used to take the analysing process to another step.


Project Category : CS / IT / Networking
------------------------------------------------------
Institute/College Name: Mait
City: Delhi
State: New Delhi
Participating Team From: Final Year

Replies

You are reading an archived discussion.

Related Posts

Project Abstract / Summary : Dump yards are the most unhealthy and the worst parts of the cities. Generally these are located at the outskirts of the cities where large...
Panasonic has recently launched its brand new smartphone, the Eluga Turbo exclusively on e-commerce site Snapdeal. The company’s latest 4G enabled device is the newest addition to the Eluga family....
Chris Tuan, professor of civil engineering at the University of Nebraska-Lincoln and his team of researchers have developed a special kind of concrete, one that can conduct electricity to melt...
Two fossil-hunting brothers in the Wales – Rob and Nick Hanigan have provided a major leap in study of the Mesozoic Era, the age of dinosaurs, by finding the remnants...
Hello Fellow Engineers, I am having trouble coming up with the correct UPS sizing for a PA System. Here are the specification. Amplifier: Battery power consumption- Idle - 0.09A ,...