Energy research is of crucial public importance but the use of computer science technologies like automatic text processing and data management for the energy domain is still rare. Employing these technologies in the energy domain will be a significant contribution to the interdisciplinary topic of ``energy informatics", just like the related progress within the interdisciplinary area of ``bioinformatics". In this paper, we present the architecture of a Web-based semantic system called EneMonIE (Energy Monitoring through Information Extraction) for monitoring up-to-date energy trends through the use of automatic, continuous, and guided information extraction from diverse types of media available on the Web. The types of media handled by the system will include online news articles, social media texts, online news videos, and open-access scholarly papers and technical reports as well as various numeric energy data made publicly available by energy organizations. The system will utilize and contribute to the energy-related ontologies and its ultimate form will comprise components for (i) text categorization, (ii) named entity recognition, (iii) temporal expression extraction, (iv) event extraction, (v) social network construction, (vi) sentiment analysis, (vii) information fusion and summarization, (viii) media interlinking, and (ix) Web-based information retrieval and visualization. Wits its diverse data sources, automatic text processing capabilities, and presentation facilities open for public use; EneMonIE will be an important source of distilled and concise information for decision-makers including energy generation, transmission, and distribution system operators, energy research centres, related investors and entrepreneurs as well as for academicians, students, other individuals interested in the pace of energy events and technologies.