Internet and mobile communication have become major potential tools in the present day societal, behavioral and technological interaction and they are creating significant impacts in countries like India and China, where the usage and application are prolific. While China has a major advantage in terms of uniform and not-so-significant language and dialect variations, India has more than 20 languages officially to contend with, besides many local dialects. This aspect has major implications in different sectors and in particular, education from primary to higher levels. With on-line courses and internet access to them being available, issues arise as to how to bring teaching and interaction to cater to different languages and dialects [1,2], and the present paper attempts to proposes a generic approach to extract information and content in different web and mobile communication modes. Without resorting to conventional data/text mining approaches, a basic pixel-based method applicable in any platform is proposed and after converting the image or text in terms of matrices, different methods to reduce the size to scalar/vector and matrices are presented. Using these as inputs, classification methods are used to extract the content and match with known patterns. The method is used for bi-lingual and multi-lingual web pages with variations and performances are detailed. The versatility of the approach in handling heterogeneous and non-structured web pages. Gottron, T (2008), Content code blurring: A new approach to content extraction, DEXA ’08: 19th International Workshop on Database and Expert Systems Applications. IEEE Computer Society, pp. 29 – 33.Gupta, S, Kaiser, G, Neistadt, D, Grimm, G (2003), DOM based content extraction of HTML documents, WWW ’03: Proceedings of the 12th International Conference on World Wide Web. New York, NY, USA: ACM Press, pp. 207– 214.