International Journal of Computer Science Issues

More than a traditional journal...

This paper describes a novel approach for addressing the issue of business-matching and recommending in business-to-business e-marketplaces. The presented matching system makes use of Collective Intelligence (CI) means for identifying and recommending business opportunities that are best correlated with the userfs need. The objective from this approach is to shift the focus from search-oriented matching, toward assistance-providing business matching systems for the next-generation e-marketplaces. At first, the content of business proposals which are submitted to the e-marketplace, along with their companies\' profiles are parsed and indexed. The indexed data may be collected from electronically submitted entries, as well as from scanned and handwritten documents. The search engine starts by expanding the queried keywords, to enable an intuitive-like search. The look-up results are then filtered based on compatibility scoring mechanisms, based on CI techniques. The personalized business-matching results and recommendations are later served to the user via a novel visual interactive graphical interface. An experimental system-prototype applying the proposed and described approach is developed and now being experimentally tested, to fully demonstrate the capabilities of the proposed system on real-world data.