Research projects of COIN are mentioned below:

 

 

Multilayer Networks/ Heterogeneous Networks: While a considerable amount of research has been done on characteristics of single networks (such as Facebook and Twitter) for over a decade, real phenomena are seldom constrained inside a single network. Indeed, many real-world systems include multiple subsystems or layers of connectivity, and it is important to consider such "multilayer" perspective to improve our understanding of these systems. When only single networks are involved, it is well known that for all the dynamical processes (such as spreading and cascading failures) the structure of the network plays an important role on the outcomes of the process. For example, behavior spreading can stall when it enters a tightly-knit community within the network. The same is true when multilayer networks are involved, but the effect of the layer structures and their interdependence may differ from the single-network case. Therefore, a great deal of interest has been recently devoted to the study of multilayer networks. In summary, there are two types of multilayer networks: (i) Multiplex Networks in which the same set of nodes are across all layers and there are multiple types of relationships between them, (ii) Interconnected (Interdependent) Networks where the nodes are of different types and such networks involve multiple networks. Multilayer networks are also known as heterogeneous, multidimensional, multiple, multisliced, multilevel networks, and networks of networks. In the COIN lab, we are focused on different research problems such as (1) Analyzing, modeling and predicting information diffusion in multilayer networks, (2) Similarity search in heterogeneous networks, (3) Network formation and Evolution, and (4) Network Visualization.

 

 

Opinion mining and sentiment analysis: The study of diverse topics such as the propagation of opinions (or its related concepts likes sentiments) about a new product over social networks, or the spreading of a virus across different species (e.g., avian flu spreading through birds and humans), requires the development of suitable spreading models that take into consideration the existence and interactions of different layers within a network. In above research domain, we are mainly focused on the roles of structural properties of a network (i.e., the topology of a multilayer network as an underlying network) in the behavior of dynamical processes over them. Other phenomena may require another approach to be considered. For example, with rapid growth of Web 2.0 technologies and social media, nowadays we have a huge volume of opinionated data recorded in digital forms such as online reviews, forum discussions, comments on blogs and social networks. In the domain of opinion mining and sentiment analysis, we are trying to process the content of a media (e.g., a tweet on Twitter) with NLP (natural language processing) methods for behavior analysis, classification, clustering and profiling. In particular, we are focused on (1) Polarity detection in user reviews on social media sites, (2) Fake review detection, and (3) Text Similarity.

 

 

Crowdsourcing: In general, crowdsourcing can be considered as using a large number of human contributors (called workers) and their capabilities for solving the problems that neither machines nor humans can solve alone. Some of more prominent applications of crowdsourcing include media (e.g., Image) tagging, item categorization, sentiment analysis, data collection, (article) writing, survey, and transcription (from image, audio and video). In addition, there are a number of online marketplaces for crowdsourcing such as Amazon's Mechanical Turk and CrowdFlower. Our work in the area of crowdsourcing is focused on quality control and aggregation methods for the answers collected from crowd workers as one of the biggest challenges in this area.

 

 

Opportunistic Networks: With the existance of various wireless technologies (such as Bluetooth and WiFi) and wireless devices with their communication and processing capabilities, Opportunistic mobile Networks emerged a mechanism of communication in wireless networks. This type of network can be used in development of IoT (Internet of Things) technology. Unlike mobile Ad hoc Networks (MANETs) in which there are end-to-end paths between nodes, Opportunistic Networks use store-carry-forward mechanism, which means the receiving nodes store the data and carry them until they can pass it to other nodes in opportunistic contacts. Opportunistic networks can be used as a means of communication in situations that there are no network infrastructures, such as natural disaster or in developing countries which broad-band communication is not widely available. By using opportunistic networks delay-tolerant data exchange is possible. In COIN, we focus on research and development of opportunistic networks as a part of IoT technology.