Information sources Search engines

Sherman, C. and G. Price. 2001. The Invisible Web Uncovering Information Sources Search Engines Can t See. Medford, NJ Information Today. Chapters give a history of the Internet and the visible Web. The contents include all kinds of searching possibilities, numerous suggestions of Web sources arranged topically, and a detailed index. 

If users are inexperienced in searching for information, they should first consult search engines, meta-databases or portals (Table 5-6). Searchers who are familiar with databases may consult known databases (numeric databases, bibliographic databases, etc.) directly, being aware that they might miss new data sources (see Section 5.18). The reliability and quality of data are only given in peer-reviewed data sources. 

ToxSeek is an NLM metasearch engine and clustering tool that enables the simultaneous searching of many different toxicology and environmental health information databases and web sites. This tool includes 59 databases including the TOXNET Search tool, as well as information sources from NLM, NIH, US Government, International and other sources. 

Science-specific search engine. Searches for different types of information (abstracts, articles, books, conferences, theses, and disssertations) in various sources (literature databases, web resources). 

The next step will help you narrow down your topic even further and make it more specific. If you asked a librarian for information, or typed John F. Kennedy into a search engine on the Internet, either the librarian or the computer would pull up thousands of sources. Most likely, you don t have the time to sift through aU the pages and books that have been written about President John F. Kennedy. For that reason, you need to refine your search. Using the 5 W s as a guide, make yourself a chart and fill in all the information that you already know. For example, your chart might look like this ... 

This set exercises concerns research on finding relevant information, as well as on preliminary organization. Start with simple search engines, then switch to more specialized search engines, and then go to the original references when necessary. Document all the sources that you have looked into and note any information you cannot find. 

Implementing semantic identification is only the first step the second is the retrieval of this information, which allows the interpretation of the relevance of the subject matter in semantic terms. Although bespoke systems can be used to do this, we really need web-wide search engines to be able to do this interpretation and use the information returned from data sources to generate hit results based on the semantic content rather than full text. 

A convenient source of information about the commercial availability of some 200,000 chemicals is M. Desing (ed.), Chem Sources—USA, published annually by Chemical Sources International, Clemson, SC 29633 in book format and as a CD-ROM. There is also a Chem Sources—Online search engine see www.chemsources.com. Many suppliers of chemicals maintain online catalogues that can be accessed on the Web. 

Many of the important sources of engineering information are subscription services. The American Chemical Society s Chemical Abstracts Service is the best source for chemical properties and reaction kinetics data. Chemical abstracts can be searched online through the SciFinder subscription service (www.cas.org). This is available in most university libraries. 

The work of scientists has been substantially eased through the availability of search engines on the Internet. A search for "preformulation" yielded 40,000 possible hits in 0.28 second for "pharmaceutical preformulation" the number was 29,000. The problems, therefore, arise regarding how to parse the data there is a lot of redundant and superfluous information available. It is for this reason that, despite the wide availability of information, knowledge must be gained from such condensed sources as this book and other books written on the subject. 

Much valuable information may be obtained by the use of the electronic information sources. For example, company annual reports, as well as lOK and lOQ reports containing a wealth of business and financial information of companies, may be retrieved on the Internet from www.investquest.com. As an alternative, the search engine AltaVista (www.altavista.digital.com) can be used to obtain similar information or to augment not only financial but also product information. Company information may be obtained by accessing a company s... 

Several lists of PEMFC mathematical models were generated from the basic search engine of the Journal of Power Sources [12] covering the 1994 to the present range. The first searches were performed to determine the yearly evolution of model appearances (search A). The second search was performed to extract more detailed information from the recent record (search B). More details are provided about these searches (completed on March 13, 2006) in Table 1. Searches A and B provided a number of papers addressing a range of topics that did not meet the initial objectives. Thns, papers addressing topics... 

Stability studies on medicines are often published in pharmaceutical journals. Other sources for stability data could be useful as well, such as the Pharmaceutical Codex [29], Kommentar zur Europaisches Arzneibuch [72] (see Sect. 39.4.8), Martindale [85] (see Sect. 39.2.4), Ophthalmika [86], Coimors [15] and Trissel s Stability of Compounded Formulations [70] (see Sect. 39.4.14). Additional information can be found in Analytical and International Pharmaceutical Abstracts. These can be retrieved by the search engine OvidSP [87]. The German collection DAC-NRF [88] (see Sect. 39.4.2) also provides much information on stability. 

There are several possible ways to retrieve relevant contextual information from the web. They all involve similar steps (1) identify reliable sources of information (2) query those sources about information related to an individual s symptom at a place and time (3) in the query results, identify that information which is most specific and relevant to the symptom, place, and time and (4) store the contextual information so that it can be linked to the original health problem. The first step must still involve manual selection by a human being the others have been automated using a search engine API such as Google Custom Search API [2,18,36]. 

Harvester collects two types of information pure text-based information and information rich in graphical elements. Text-based information is retrieved and indexed from the following public databases and prediction servers UNIPROT, SOURCE, SMART, SOSUI, PSORT, RZPD, Homo-logene, gfp-cDNA, IPI, CD ART, STRING. For optimal search engine indexing, redundant text information is removed by server-specific converter modules. 

---