The mission of the Journal of Machine Learning Gossip (JMLG) is to provide an archival source of important information that is often discussed informally at conferences but is rarely, if ever, written down.
The mission of the Journal of Machine Learning Gossip (JMLG) is to provide an archival source of important information that is often discussed informally at conferences but is rarely, if ever, written down.
Mainly based on implementing a router infrastructure which is independent from outer sources (it is working locally) and it transmits data to the mobile phones or laptops which are simply depicted as receiver devices, they don’t send anything back, basically the whole idea is that the routers do the whole work of sending news items to the mobile devices and the gossip algorithm is implemented between the routers only, they also got implemented a shuffle function which sends a random item to other routers and is removing it from its current accumulation of items so that there is a higher probability of someone getting it even if he wasn’t passing nearby that specific router. The mobile devices can filter out news that they don’t want to receive. If there is a duplicate of an item in the router’s storage then it is being erased. There can and should exist replicas of items in other router’s storage.
In this solution there is an addition of synchronization functionality which prevents loss of publications during the distribution of subscriptions among the nodes.
R. Baraglia, P. Dazzi, M. Mordacchini, and L. Ricci. Journal of Computer and System Sciences, 79 (2):
291 - 308(2013)<ce:title>10th IEEE International Conference on Computer and Information Technology, 2010</ce:title>.
S. Korada, A. Montanari, and S. Oh. Proceedings of the ACM SIGMETRICS joint international conference on Measurement and modeling of computer systems, page 209--220. New York, NY, USA, ACM, (2011)