期刊名称：International Journal of Computers and Communications
印刷版ISSN：2074-1294
出版年度：2015
卷号：9
页码：44-55
出版社：University Press
摘要：Wouldn't it be nice to say or type "show me all the pictures from last Saturday party" on a browser and get all the requested pictures from Facebook? Or do specific transaction like "I would like to buy the following book - Artificial Intelligence by Stuart Russel, 3rd edition; please use my credit card on file and ship it to my home address (assuming that the user is already logged on to the specific website)" and receive the requested book on time? Or ask a question and get a specific answer? Or get summary of an article? Or get a much better prediction from a BI (Business Intelligence) or Analytics software? In fact, based on good research, we see a clear trend that the future Internet will be something that can provide very specific, more precise and direct information (like the examples mentioned above) in a very easy way so that anyone including an illiterate person can access and use it at ease. We call this Intelligent Internet (IINT). Existing search engines usually provide thousands to millions of search results for any typical search. It is not easy even for advanced users to find the desired results from such a large set. One cannot get a specific answer or a set of answers to a question typed in a search engine. There is no automated way to get a good summary of a document or get a good inference from a document. Similarly, there is no way to get some specific desired information like "basic information of last 3 flights I took on United Airlines". However, as mentioned, these are the key features that users would expect from next generation internet. Moreover, users would like to use such features in a natural way - like using a natural language sentence (by typing or preferably, by saying it; and for many cases using sentences that may not be grammatically correct). This is obviously a very complex task (and hence not solved yet). We would need multiple approaches, algorithms and technologies to achieve these. For example, Natural Language understanding (NLU), Big Data and Intelligent Agent (IA) are the 3 key areas we need to focus on. A Semantic Engine is the core engine that is needed for all these 3 major areas. In this paper we describe how a Semantic Engine using Brain-Like Approach (SEBLA) can address all key complexities of the next generation Internet and effectively provide all the key desired features mentioned above. We focus on Intelligent Search, Question Answering (Q & A System) and Summarization. We also show how NLU, IA and Big Data can be integrated with existing client-server based web application architecture using the design patterns (e.g. Model-View-Controller) and software frameworks (e.g. Java or Ruby-on-Rails).
关键词：Big Data; Natural Language Understanding (NLU); Semantics; Artificial Intelligence; Intelligent Internet; Intelligent Search; Question & Answer System; Summarization; Knowledge Extraction; Intelligent Agent; Machine Learning; Predictive Analysis; Business Intelligence; Information Technology.
