OSTI recently launched SciTech Connect, a portal to free, publicly available DOE research and development (R&D) results. SciTech Connect incorporates the contents of two of the most popular core DOE collections, DOE Information Bridge and Energy Citations Database, and employs an innovative semantic search tool enabling scientists, researchers and the scientifically attentive public to retrieve more relevant information. Other features include faceting, in-document search, word clouds, and personalization. There are over 2.5 million citations, including citations to 1.4 million journal articles, 364,000 of which have digital object identifiers (DOIs) linking to full-text articles on publishers’ websites. SciTech Connect also has over 313,000 full-text DOE-sponsored scientific and technical reports; most of these are post-1991, but close to 85,000 of the reports were published prior to 1990.
OSTI will gradually phase out its current DOE Information Bridge and Energy Citations Database products and replace them with the improved search interface of SciTech Connect (read more about the transition details and scope of the new product). Consolidated in SciTech Connect, DOE Information Bridge and Energy Citations Database accounted for approximately half of the 298 million transactions OSTI handled in 2012. OSTI will work to ensure a smooth transition for patrons as it consolidates these two web-based services into SciTech Connect.
SciTech Connect employs a semantic search technique known as keyword-to-concept mapping. This means that SciTech Connect accepts keyword-based queries and returns concept-mapped queries as in a taxonomy; a search term is mapped to other associated terms, including narrower and related concepts.
In this way, semantic search enables the new SciTech Connect search engine to recognize and make use of the logical relations among concepts in different scientific documents, regardless of whether those documents use standard descriptors to express those concepts. As a consequence, even the casual user easily recognizes the superiority of semantic search results over traditional word/phrase search results in a side-by-side comparison (see the Search Tip comparison below).
The keyword-to-concept mapping used by SciTech Connect is just one of many different ways to perform semantic searching. In general, semantic search is a way to enhance search accuracy contextually. Rather than relying on search algorithms that identify a specific query term, semantic search uses more complex contextual relationships among people, places and things. It is an especially effective search approach when a person truly is researching a topic, rather than trying to navigate to a particular destination.
Users of SciTech Connect will find that their search results are somewhat different when using the Basic Search versus the Advanced Search. The Basic Search offers a semantic technique. This means the Basic Search adds related terms to your search. Advanced Search does not use the semantic technique but does use Boolean logic. Therefore, users might find a significantly higher number of citations in the search results from the Basic Search. This is because when you search a term or phrase in Basic Search, you will get citations with your term or phrase, and you will also get citations for narrower and related terms. If you are searching a specific term or multi-term phrase and want only citations with that specific term or phrase, then you should use the Advanced Search. Compare the results of a search for "Greenhouse Effect" in Basic Search and Advanced Search. Remember, when searching a multi-term phrase, such as "Climate Change" or "Enriched Uranium," you should put the terms in quotes regardless of whether you are using Basic or Advanced Search. You can clearly see the difference in these Basic Searches for Enriched Uranium with quotes and without quotes.