Business objectives are accomplished successfully when human resource management systems are developed and implemented according to organizational goals, particularly if personal information management (PIM) is adopted to utilize all employee information productively. Some PIM systems use Semantic Desktops, the semantic personal desktop layer for integrating applications and personal life items, as the means to support users in information management. More and more organizations/enterprises are taking advantage of mashups, which support users in fast integration of heterogeneous data from multiple sources. However, although most people and organizations/enterprises benefit from the collaborative principles of Web 2.0 technologies by using social networking sites (SNSs) to build their social activities/relations and support their knowledge management, the unstructured data overload is increasing and distributing in both human lifetime and workplace. In addition, those PIM systems are limited to local storage or isolated data repositories, and do not fulfill most of the requirements for a collaborative environment, above all at the organizational level. Therefore, a key issue arises in the necessity to provide a flexible and semantic-based way for bringing in internal and external data sources (especially personal information sources from Semantic Desktops and SNSs) into enterprises. This thesis aims to utilize semantic web and mashup technologies for semantic-based information integration and to leverage existing personal resources in Semantic Desktops and SNSs. A lightweight mashup language and a semantic-based mashup framework are proposed to enable a semantic-aware mashup dataflow that primarily supports non-experts to create mashup data for personal/organizational use. In addition, reusable components for Web 2.0 information retrieval are developed to prepare mashable resources and trustworthy mashup data. Subsequently, the mashup results can be combined with other policies for self-monitoring purposes in preventing the disclosure of personal/organizational information in Web 2.0 via SNSs. The introduced mashup system and its components could be adapted to the layered approach of Open Semantic Enterprise for the semantic data integration in enterprises. Such adaption would support knowledge workers and enable activity-oriented collaboration, as well as to team up with coworkers in the collaborative environment of enterprises.