Abstract
Social networking have revolutionized traditional information-sharing methods. They are mostly built on an individual’s offline social circle and they provide users with a wide variety of virtual interaction mechanisms. The purpose of this research is to assess the intensity of this problem by identifying SNSs users’ personal information disclosure levels, the kinds of information that they reveal, the degree to which they expose personal information to the public, the privacy settings they apply and their level of knowledge and awareness about how the data they share is protected by SNSs service providers. Primary and secondary methods of data collection were used. Primary data were used to collect information about users’ behaviour on SNSs. The research focused on two different social networks: Twitter and Instagram.The results showed that there was a significant amount of disclosure of personal information; however, it differed from one social network to another.The study concludes by offering recommendations and guidelines that may provide a safer browsing experience for social network users in Nigeria.
CHAPTER ONE
INTRODUCTION
1.1 Background To The Study
In social media analysis, measuring different structural properties of a social network can help better understand individuals and their roles within the large network. For example, in determining which nodes are most important or influential in the social network, one need to define measures for quantifying centrality, level of interactions, and similarity, among other qualities. To compute these measures, a graph representation of a social interaction is taken in as input, such as nodes friendships (adjacency matrix), from which the measure value is computed. By using graph measures of centrality we can identify the most prominent actors commonly known as the key players in the network. In modeling and mining social media several centrality measures are defined among them degree centrality which describes nodes degree which is the number of edges the node has whereby if a given node has high degree, the more central the node is. By using these measures, one can identify various types of central nodes in a network and answer questions like “Who are the influential individuals in the network?”. The second measure is closeness centrality which indicates how close a node is to all other nodes in the network whereby rather than considering the neighbors node, all nodes are taken into consideration (Magnusson 2012). It indicates nodes as more central if they are closer to most of the nodes in the graph and it has measured as the average distance from the source vertex to any other vertex within the graph. This metric allows us answers questions like “What interaction patterns are common in within friends?”. The third metric between-ness centrality indicates how important a node is to the shortest paths through the network and measures to what degree an actor has to traverse through a specific node in so as to reach other nodes within the network. The last measure is the Eigenvector centrality which unlike centrality measures, it tries to generalize degree centrality by incorporating the importance of the neighbors and the influence an actor has in the social network (Golbeck 2013).
However, the emergence of social networking sites (SNSs) in people’s daily lives has transformed the way users communicate and share information. Before SNSs, people’s means of communication and information sharing were very limited, especially in terms of interaction, and people mostly communicated with others they knew personally. Currently, individuals use SNSs to share user-generated content online via computers or smartphones in many different formats, depending on the social network of their choice (Ge, Peng, & Chen, 2014). Users share news about their lives effortlessly, whether it is in the form of a video, a photo, a post or a status update. In addition, users nowadays share information with a much larger audience, sometimes larger than they intend.
Advances in technology have enabled SNSs to develop enormously in a way that has created new methods of sharing information. Social networks began as websites where users only had access via a laptop or desktop. However, with the development of smartphones, social networks released mobile application versions and/or developed stand- alone mobile applications. This development made it easier and more convenient for users to access their online profiles, updating them more actively and in real time (Aldhafferi, Watson, & Sajeev, 2013). However, the more accessible the social network and the easier it is to use, the more information users share (Coyle & Vaughn, 2008) due to its constant presence in their lives. SNSs unquestionably have a strong social impact; however, as a result, the lines between individuals’ virtual and offline lives have been blurred.
As of August 2016, there were over 2.34 billion social network users globally. This number is expected to increase to 2.95 billion social networks users by 2020, which is approximately a third of the world’s entire population (Statista, 2016). Due to the increased use of SNSs, social networks have become rich sources of users’ personal information. Users’ personal information is very valuable to many different parties and can be exploited for financial gain. Firstly, advertisers can invade users’ privacy by accessing their personal information and browsing habits, which is supplied by SNS providers, in order to recommend products and services; such promotions are referred to as targeted personalised ads. Secondly, sharing of personal information such as full name, age, gender, and other personal information such as family photos leaves users vulnerable to online criminals, who may exploit such information for malicious actions such as identity theft or online stalking. Such actions can affect users’ safety and cause not just financial loss but also emotional distress to the victims. The following are the three main research questions this research aims to answer.
1.2 Statement of the Problem
With the proliferation of digital technology, Internet of Things and smart devices, large data now stream every day from social networks, mobile phones, credit cards and computers, city infrastructure and sensors networks among others. This data has grown exponentially resulting to what is commonly known as “Big Data”. With this ever-increasing volume of data, forensic analyst faces challenges in investigations involving huge data volumes from social networking sites while at the same time limited by computer processor speed, memory and storage resources of a single node.
The traditional digital forensic tools focus on transactional data commonly known as structured data for analysis in a relational or hierarchical database. Most widely used traditional forensic tools have not undergone any major architectural change (Edwards 2011) hence lacks suitable features to handle big data forensic investigation. Again, the use of traditional tool to analysis Big Data is time consuming, resources intensive and correlation of evidence from multiple source is not feasible. The ability to derive insights and correlate artifacts found is such big data become difficult using the traditional forensic tools. The range of data from social network sites for forensics increases considerably and increases further with numerous participants involved in social media resulting into challenges in carrying forensics investigation involving these large volumes of data. With this increased social network data, it has become difficult to collect, store and analyze such big data on a single computer node.
In order to collect, store and analyze such data fast and effectively, Apache Spark a leading distributed computing framework come in handy with features that can process voluminous amount of data that can range from terabytes to petabytes of data. Forensic analysis of social networks can help law enforcers understand and solve various cybersecurity problems, including uncovering the social media cybercrimes. The large data and the complex structure of social network sites calls for research and design of new generation of forensics analytics methods and tools that can effectively process and correlate digital evidence found in big data more often in real-time basis. The study developed an apache spark based big data forensics tool for social networks cybercrime detection and systematically analyzed big data from Twitter social network site to identify, collect, preserve and analysis artifacts that are relevant in Twitter social network forensics to supplement the shortcoming of traditional forensic tools in carrying out large data forensic investigations.
1.3 Aims and Objectives of the Study
The main objectives of this study is to investigate Social Network Analysis: Effect of Data Over sharing
The specific objectives are :
- To examine the personal attributes that can have an influence on information disclosure and the privacy settings of social network users
- To investigate the personal attributes that can have an influence on information disclosure by and the privacy settings of social network users
1.4 Research Questions
Q1: What are the personal attributes that can have an influence on information disclosure by and the privacy settings of SNS users?
Q2: How do users’ levels of privacy concern affect the amount of information they disclose in social networking sites1.5 Scope of the Study
There are several social networks in existence today including Twitter, WhatsApp, Myspace, Facebook, LinkedIn and Instagram among others. It will not be possible to carry out the study on all social network sites due to time and resource constraints, and therefore twitter and instagram Will be use
