Alchemy includes one general setting when using the web interface that will affect the search results. Changes in the settings are unique to the end-user conducting the search.
General settings that affect Search
Access the general settings from the web interface by selecting the Search tab then selecting the Settings folder.
In the Word Matching section, select the Rank search results check box to rank documents in the search results list by relevancy.
Note: Alchemy does not support multiple field fuzzy searches. When you want to perform a fuzzy search, you must perform the search one field at a time.
About ranking search results This intelligent ranking is based on both the term count and the inverse document frequency. Ranking based on term count functions as follows: a document that contains all of the search words entered is ranked higher than a document that only contains one or more but not all of the words entered (this is shown by a percent-age value). Inverse document frequency is used in conjunction with the term count logic to further rank documents. Those search terms that occur relatively infrequently in the entire database (or group of databases) are weighted higher in importance than those search terms that occur frequently throughout the database. For example, say a user is searching for the words imaging archival Alchemy. In a database that contains imaging trade magazine articles, by inverse document frequency the search word imaging will carry less weight than the word Alchemy simply because imaging occurs more frequently. Furthermore, an article that contains many instances of the word Alchemy along with instances of the words imaging and archival will likely be ranked high in the results list.
About fuzzy word matching When you rank search results, you can set the word matching level, between exact and fuzzy, for the words you type in the fields on the Search dialog box. An exact search will locate exact duplicates of search words, while a fuzzy search uses pairs of characters from the search word to locate (match) similar pairs. For example, an exact search on the word raincoat will locate duplicate words. A fuzzy search on the word raincoat will examine and locate words containing the same matching pairs found in the search word raincoat: ra ai in nc co oa at Use the slide bar under the Rank search results check box to select the desired degree of fuzzy searching. The degree of search settings are set to 10% increments, with a full fuzzy search requiring a 50% match of the pairs found in a word. Fuzzy searching is most successful with longer words (five or more characters); use wildcards for words containing four or less characters. Fuzzy searching will add more search words to the query, resulting in a larger results list. Fuzzy searching will not perform semantic searching (e.g., searching for the word water will not find the word steam). 2. If desired, change the default operator to use between fields by selecting or in the Boolean operator to use between fields section.
BOOLEAN SEARCHES:
Boolean searches allow you to search for multiple terms in the same field. The can be either AND or OR. The selection made on this screen will be unique to the user.
SAVED SEARCHES:
Common and repetitive searches can saved and retrieved by the user. Once all the criteria has been entered the user can select the "Saved Search" icon, noted with the arrow, and enter a name for the criteria used. As searches are saved they will appear in the drop down list as shown above and be easily selected the the user.
Recent Comments