CS 224 W Final Report Group 37
نویسندگان
چکیده
Much of the current research is being done on social networks, where the cost of an edge is almost nothing; the click of a button in most cases. However, in a graph where each node is an investor and/or investee and each edge is an investment in a company, each edge costs thousands, millions, or potentially billions of dollars. On top of that, each edge is carefully considered, calculated, and subject to a large set of laws, regulation, prior affiliations, and so forth. And most importantly, each edge is intended to ultimately make money for the investor. Investment networks also have at least two primary node types: investors and investees. Edges would mostly occur between investors and investees (see below for a description of why this is occasionally not true), leading to a nearly bipartite structure which can complicate the network structure for certain tasks. It is possible to separate nodes even further by looking at the kinds of investors that occur in the graph. In this work, we look at the CrunchBase investment network and compare this network to previously studied social/information networks, investigate the structure induced by the variations in node types, and then attempt the difficult task of investment (link) prediction on the network.
منابع مشابه
CS 224 W Project Final Report CUDA Implementation of Large Graph Algorithms Group
Running SCC graph algorithms on large datasets can be a time-consuming task, and we spent the quarter investigating methods of parallelizing this task using CUDA. For very large graphs, too much time can be wasted by not parallelizing the graph algorithms, and we want some of the insights from our experiments to be used to speed up common graph analysis tasks. We initially started by implementi...
متن کاملCS Study Group FY 07 Phase 2 Machine - Checked Metatheory for Security - Oriented Languages Final Technical Report
متن کامل
Autumn CS 224 w Social and Information Network Analysis Final
User generated online reviews on products or services are valuable to other users to make informed decisions. Thus, identifying experts in reviewers becomes important. We tackle the problem of reviewer expertise ranking in online review datasets by analyzing the bipartite graphs of reviewers and reviewed items, an approach that has not been considered in previous studies of the problem. We use ...
متن کاملUsing a Caesarean Section Classification System based on characteristics of the population as a way of monitoring obstetric practice
OBJECTIVE to compare the distribution of caesarean rates in the Robson's 10 groups classification in order to see if any change occurred after the implementation of an audit and feedback intervention. DESIGN cross sectional, before and after an audit and feedback study. SETTING a university hospital in Brazil. METHODS clinical records of all births during two three months-periods were eva...
متن کاملPatent Co - Authorship Network Analysis CS 224 W Final Report
This project aims to model the h-index of a firm as a function of several properties of the owning organization’s patent co-authorship and citation networks. Studies have shown that the number of forward citations a patent receives strongly correlate with its market value [3], which is of great interest to potential investors. Unfortunately, future number of forward citations of particularly hi...
متن کامل