الفهرس 
English Cover
ContentsOnly 14 pages are availabe for public view
 |  | from | 69 |  |  |
| | | from | 69 | | |
Abstract
With the advent of the big-data era and the rapid growth of the amount of data, companies are faced with more opportunities and challenges to outperform their peers. Utilizing the full bene t of social media requires companies to identify their customers against customers as a whole by linking their local data against data from social media applying some matching techniques that di er from simple to complex on what is called record linkage. Record linkage is the process of grouping together related records that are believed to belong to the same entity which leads to richer information and avail hidden business facts and opportunities.We applied a particular kind of a graph record linkage based on the interactions between entities (Block rank) between objects in both the data sources and calculated the similarity score between entities based on string similarity and network similarity to e ciently achieve better linkage results.Our objective is to match customers’ pro les across social media and local cooperate by applying both string and network-based similarity.This study serves new business opportunities like de ning crucial top in uencer’s customers, churn reasons, customer behavior and customer segmentation that leads to increasing customer loyalty and satisfaction and creates a new channels and ways to provide tailored promotions to retain customers and decrease churn rates for the corporate in the telecommunication domain which has been initiated with the analysis of customers’ behaviors over social media as part of big data implementation project on Vodafone Egypt