Social network analysis tools facilitate qualitative or quantitative analysis of social network by describing network’s feature either via visual or numerical representation. It generally uses network or graph theory to examine social structures. The main components are nodes (people) and the edges that connect them. Some of them performs predictive analysis too. Below, we have listed some of the most effective social network analysis software that are available for free.
AllegroGraph is a graph database developed to store RDF triples. It is fully transactional OLTP database, which stores data structured in graphs rather than in tables. It includes a social networking analytics library, and storage component for the TwitLogic project that aims to bring the semantic web to twitter data.
Commetrix is a dynamic network visualization and analysis software that provides easy exploratory yet comprehensive access to network data. It creates a rich expert network map, recommendation systems from communication logs, and focuses on analyzing evolving patterns of electronic communication like email, voice over IP and instant messaging.
Socilab is an online tool that lets you visualize and analyze LinkedIn network using methods derived from social-scientific research. It displays a number of network measures drawn from sociological research on professional networks, and percentile bars comparing your aggregate network measures to past users. Also, there is a messaging feature that allows you to type and send a message to the selected LinkedIn contacts.
Cuttlefish is a network workbench application that allows visualization of network data, graph edition, interactive manipulation of the layout, and process visualization.
To represent network data, the tool uses Cuttlefish eXtended Format that defines network data in terms of edges, vertices, weights and visual information like shapes, color and label. The other file format is Cuttlefish Evolution Format that defines the changes happening in a network. It also supports older file formats, including GraphML and Pajek.
18. Social Network Visualizer
This is cross-platform user friendly tool that allows you to draw social network with a few clicks on a virtual canvas. Either load field data from a file (in supported format) or crawl the internet to create a social network of connected webpages.
Social Network Visualizer calculates standard graph and network cohesion metrics (like density, eccentricity, clustering coefficient, etc.), matrix routines, and centrality and prestige indices. Moreover, it supports fast algorithms for community detection, structural equivalence analysis, multirelational network loading and editing and random network creation using various random network generation models.
JUNG stands for Java Universal Network/Graph Framework. This Java application provides an extendible language for the analysis, modeling and visualization of data that could be represented as a graph or network.
JUNG supports numerous graph types (including hypergraphs) with any properties. It enables customizable visualizations, and includes algorithms from graph theory, social network analysis and data mining. However, it is limited by the amount of memory allocated to Java.
Tulip is dedicated to the analysis and visualization of relational data. It enables the development of algorithms, interaction techniques, domain-specific visualization, visual encodings and data models. It also allows reuse of components which makes the framework efficient for research prototyping and end-user application development.
Statnet is a suite of R packages that perform a wide range of data management, visualization and statistical network analysis tasks. This analytic framework is based on Exponential family Random Graph Model and provides tools for model estimation, evaluation, simulation and network visualization. Moreover, the statistical modeling include dynamic and cross sectional modeling, latent space and latent cluster models.
Netlytic is a cloud-based text analyzer and social network visualizer that can automatically summarize large dataset of text and visualize social networks from conversations on social media sites like Twitter, YouTube, online forums, and blog comments. The tool is mainly developed for researchers to identify key and influential constituents, and discover how information flow in a network.
NetworkX is a Python package for creating, manipulating, and study the structure of dynamics, and functions of complex networks. It includes many algorithms, metrics and graph generators. The tool is capable to construct random graphs incrementally, and capable to find cliques, subgraphs and k-cores. Furthermore, it can explore adjacency, degree, diameter, center, radius, and draw networks in 3 dimensions.
Cytoscape is used for visualizing complex network and integrating these with any type of attribute data. The tool is very vast in features – it lets you customize network data display, filter the network to select subsets of nodes, search target nodes and edges, and layout the network in two dimensions from different network layout algorithms including cyclic, tree, edge-weight, force-directed, and more.
Subdue discovers structural, relational patterns in data representing entities and relationships. It uses minimum description length methodology to identify patterns that diminishes the number of bits required to describe the input graph after being compressed by the pattern.
Subdue can also perform numerous learning task, such as supervised and non supervised learning, clustering and graph grammar learning. Apart from social network analysis, it has been successfully applied in Bioinformatics, counter terrorism, aviation and web structure mining.
This graph visualization software represents structural information as diagram of abstract graphs and networks. Graphviz has many graph layout programs suitable for social network visualization. It takes description of graphs in a simple text language, and creates diagrams in useful formats, like PDF for inclusion in other documents, display in an interactive graph browser or SVG for web pages.
Moreover, it has different helpful features for concrete diagrams, for instance, options for fonts, color, line styles, tabular node layout, custom shapes and hyperlinks.
NetMiner comes with 14 days trial period. It is used for analysis and visualization of vast network data based on social network analysis. The features like data transformation, visualization of network data, chart and Python script language help you detect underlying patterns and structures of the network.
SocioViz is a social media analytics platform for digital journalists, social researchers and media marketers. It lets you analyze any topic, term or hashtag, identify key influencers, opinions and contents and export the data in Gephi format for further analysis.
UNISoN is a Java application that can analyze messages to save to a Pajek-format file for social network analysis. It generates networks using the author of each post. If someone interacts with a post, a unidirectional link is created from the post’s author to the author of the message they are replying to. Also, there is a preview panel that displays the network visually.
NetworKit is a growing platform for large-scale network analysis, in range form thousands to billion of edges. It implements efficient graph algorithm, most of them are parallel to utilize multicore architecture. They are supposed to calculate standard measures of network analysis like clustering coefficients, degree sequences, and centrality measures. Furthermore, it aims to support a variety of input and output formats.
GraphStream is designed for the modeling and analysis of dynamic graphs. It lets you create, import, export, shape and visualize them. Instead of only a set of edges and nodes, graphs are defined as a “flow of graph events”. Events tells when an edge, node or associated component changes. Thus, a graph is not described as fixed representation, but by the entire evolving history of graph elements.
NodeXL is an open source template for Microsoft Excel for network analysis and visualization. It allows you to enter a network edge list in a worksheet, click a button and visualize your graph, all in the familiar environment of the Excel window.
The tool supports extracting email, YouTube, Facebook. Twitter, WWW, and Flickr social network. You can easily manipulate and filter underlying data in spreadsheet format.
R programming language is packed with numerous packages relevant for social network analysis – igraph for generic network analysis, network for manipulating and displaying network objects, sna for performing sociometric analysis, tnet for performing analysis of weighted or longitudinal network, Bergm for Bayesian analysis for exponential random graph models, networksis for simulating bipartite networks with fixed marginals, and many more.
Gephi is usually a graph exploration and manipulation software written in Java. It provides an easy way to create social data connectors to map community organizations and small-world network. Along with social network analysis, it performs exploratory data and link analysis, and biological network analysis. Perhaps the most advanced free analysis tool.
The software helps you to explore and understand graphs. You can interact with the figures, manipulate the structures, color and shape to reveal hidden properties. The multi-task and flexible architecture lets you work with complex data and produce valuable visual results. Moreover, there is a 3d render engine capable of displaying large networks in real-time, just to speed up the exploration.
Pajek is used for analysis and visualization of large networks containing up to one billion vertices. The program performs this by using six data types – graphs, vertices, vector (properties of vertices), cluster (subset of vertices), permutation (reordering of vertices) and hierarchy (general tree structure on vertices).
Pajek is not a “one click program”; for getting results several basic operations must be executed in a sequence. Some of the basic operations include, shrinking specific part of networks, searching for connected components, searching for shortest paths, maximum flow, k-neighbors, centralization of networks, fast sparse network multiplication, generating different types of random networks and many, many others.