Results 1 -
6 of
6
Influence at Scale: Distributed Computation of Complex Contagion in Networks
"... We consider the task of evaluating the spread of influence in large networks in the well-studied independent cascade model. We describe a novel sampling approach that can be used to design scalable algorithms with provable perfor-mance guarantees. These algorithms can be implemented in distributed c ..."
Abstract
-
Cited by 1 (0 self)
- Add to MetaCart
(Show Context)
We consider the task of evaluating the spread of influence in large networks in the well-studied independent cascade model. We describe a novel sampling approach that can be used to design scalable algorithms with provable perfor-mance guarantees. These algorithms can be implemented in distributed computation frameworks such as MapReduce. We complement these results with a lower bound on the query complexity of influence estimation in this model. We validate the performance of these algorithms through exper-iments that demonstrate the efficacy of our methods and related heuristics. 1.
SEISMIC: A Self-Exciting Point Process Model for Predicting Tweet Popularity
"... Social networking websites allow users to create and share content. Big information cascades of post resharing can form as users of these sites reshare others ’ posts with their friends and followers. One of the central challenges in understanding such cascading be-haviors is in forecasting informat ..."
Abstract
-
Cited by 1 (0 self)
- Add to MetaCart
(Show Context)
Social networking websites allow users to create and share content. Big information cascades of post resharing can form as users of these sites reshare others ’ posts with their friends and followers. One of the central challenges in understanding such cascading be-haviors is in forecasting information outbreaks, where a single post becomes widely popular by being reshared by many users. In this paper, we focus on predicting the final number of reshares of a given post. We build on the theory of self-exciting point pro-cesses to develop a statistical model that allows us to make accu-rate predictions. Our model requires no training or expensive fea-ture engineering. It results in a simple and efficiently computable formula that allows us to answer questions, in real-time, such as: Given a post’s resharing history so far, what is our current estimate of its final number of reshares? Is the post resharing cascade past the initial stage of explosive growth? And, which posts will be the most reshared in the future? We validate our model using one month of complete Twitter data and demonstrate a strong improvement in predictive accuracy over existing approaches. Our model gives only 15 % relative error in predicting final size of an average information cascade after ob-serving it for just one hour.
Hadi Daneshmand 1/3 RESEARCH STATEMENT
"... Diffusion of virus in computer networks, information cascade in social networks, cascade failure on networks, diffusion of diseases between peo-ples are samples of diffusion on complex networks. Actually, diffusion or cascade is a general term for transmission trough links which is established among ..."
Abstract
- Add to MetaCart
(Show Context)
Diffusion of virus in computer networks, information cascade in social networks, cascade failure on networks, diffusion of diseases between peo-ples are samples of diffusion on complex networks. Actually, diffusion or cascade is a general term for transmission trough links which is established among persons or objects. Mathematical modeling of this transmission pro-cess provides a cascade model which is distinctive to the context and type of the network. In particular, continuous time diffusion model [5] is a dif-fusion model that proposed to model diffusion process in a social network. In fact, my research course lies in inference network structure by time of infection based on continuous time diffusion model. In some applications network structure is not available and only infection time of network nodes are available. For instance, in viral marketing we have only time of pur-chasing a product. Furthermore, peoples don’t mention reference of posts in many information sources such as web logs [8]. Link Prediction in Signed Social Networks Based on Interaction of
Learning and Controlling Network Diffusion in Dependent Cascade Models
"... Abstract—Diffusion processes have increasingly been used to represent flow of ideas, traffic and diseases in networks. Learning and controlling the diffusion dynamics through management actions has been studied extensively in the context of independent cascade models, where diffusion on outgoing edg ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract—Diffusion processes have increasingly been used to represent flow of ideas, traffic and diseases in networks. Learning and controlling the diffusion dynamics through management actions has been studied extensively in the context of independent cascade models, where diffusion on outgoing edges from a node are independent of each other. Our work, in contrast, addresses (a) learning diffusion dynamics parameters and (b) taking management actions to alter the diffusion dynamics to achieve a desired outcome in dependent cascade models. A key characteristic of such dependent cascade models is the flow preservation at all nodes in the network. For example, traffic and people flow is preserved at each network node. As a case study, we address learning visitor mobility pattern at a theme park based on observed historical wait times at individual attractions, and use the learned model to plan management actions that reduce wait time at attractions. We test on real-world data from a theme park in Singapore and show that our learning approach can achieve an accuracy close to 80 % for popular attractions, and the decision support algorithm can provide about 10-20 % reduction in wait time. I.
Back to the Past: Source Identification in Diffusion Networks from Partially Observed Cascades
"... When a piece of malicious information becomes rampant in an information diffusion network, can we identify the source node that originally intro-duced the piece into the network and infer the time when it initiated this? Being able to do so is critical for curtailing the spread of malicious informat ..."
Abstract
- Add to MetaCart
When a piece of malicious information becomes rampant in an information diffusion network, can we identify the source node that originally intro-duced the piece into the network and infer the time when it initiated this? Being able to do so is critical for curtailing the spread of malicious information, and reducing the potential losses in-curred. This is a very challenging problem since typically only incomplete traces are observed and we need to unroll the incomplete traces into the past in order to pinpoint the source. In this pa-per, we tackle this problem by developing a two-stage framework, which first learns a continuous-time diffusion network model based on historical diffusion traces and then identifies the source of an incomplete diffusion trace by maximizing the likelihood of the trace under the learned model. Experiments on both large synthetic and real-world data show that our framework can effec-tively “go back to the past”, and pinpoint the source node and its initiation time significantly more accurately than previous state-of-the-arts. 1
MPI for Intelligent Systems
"... Information spreads across social and technological networks, but often the network structures are hidden and we only observe the traces left by the diffusion processes, called cascades. It is known that, under a popular continuous-time diffusion model, as long as the model parameters satisfy a natu ..."
Abstract
- Add to MetaCart
Information spreads across social and technological networks, but often the network structures are hidden and we only observe the traces left by the diffusion processes, called cascades. It is known that, under a popular continuous-time diffusion model, as long as the model parameters satisfy a natural incoherence condition, it is possible to recover the correct network structure with high probability if we observe O(d3 logN) cascades, where d is the maximum number of parents of a node and N is the total number of nodes. However, the incoherence condition depends, in a non-trivial way, on the source (node) distribution of the cascades, which is typically unknown. Our open problem is whether it is possible to design an active algorithm which samples the source locations in a sequential manner and achieves the same or even better sample complexity, e.g., o(d3i logN), than previous work. 1.
