Task: Compare 2 special search sites or tools of interest and discover which one you think works better than the other.
Note: Click to enlarge images
Google Squared and Wolfram Alpha are similar in the sense that they are both special search tools that attempt to acquire and extract structured data. They both have simple and similar search interfaces (refer to the images above) in which the search box is the predominant object and is centred on the page for easy and direct access.
Even though they were both created for similar purposes, their approaches in accomplishing this challenging task are very different from each other.
These differences include:
- Is a ‘Search Engine’ that extracts its results from across the web
- Presents its results in spreadsheet-like format where each valid query returns a table of search results which has its own set of columns ( refer to figure 1.C)
- Allows adding a column to help with the search query
- Can return images with the search results if available
- Returns numerous results focusing on recall rather than precision
Recall = # number of relevant documents retrieved / # relevant documents in collection (Morville & Rosenfeld, 2007).
- Is a ‘Computational Knowledge Engine’ that generates output by doing computations from its own internal knowledge base
- Presents its results in a report-like format with headings and subheadings (refer to figure 1.D)
- Can return informative graphs with the search result
- Returns specific data focusing on precision rather than recall
Precision = # relevant documents retrieved / # total documents in collection (Morville & Rosenfeld, 2007).
It’s important to note that Google Squared has gone with a more ‘descriptive content component’ to present their results. This is for users who are still quite unclear in what they are searching for, in which case, the results can return a summary, part of an abstract or keywords to help the user with their query which can be seen in the description column in figure 1.C.
Wolfram Alpha on the other hand uses more of a ‘representational content component’ to present their results. This style is more suited to this site as the results contain facts specific to the query. As a consequence the user should have a clear understanding in what they’re searching for.
A major part of any search tool or engine is to make clear the sources in which the data derives from. Both these sites provide a source list in which the user can view and click to go to the actual page in which the information was taken from (refer to figure 1.E & 1.F).
Both these sites also contain various query builders that assist in its query performance.
Google Squared: Stemming tools, natural language processing tools and controlled vocabularies and thesauri.
Wolfram Alpha: Spell-checker, stemming tools, natural language processing tools, math calculations, Morse code, and many more.
Refer to the link to see all the options: http://www.wolframalpha.com/examples/
As for which site works better than the other? It purely depends on the type of search that the user is performing, whether they are after recall (Google Squared) or precision (Wolfram Alpha) results. Both sites have their pros and cons and neither one can cater for everyone’s needs. But instead, they cater for a specific audience in mind in which they can best provide for.
Computer Scientist Calls for Natural Language Q&A Searching
Q&A Search: Who, What, Where, When, Why & How
Understanding the web to find short answers and “something different”
What is Wolfram Alpha good for?
Wolfram Alpha Is Coming!
Morville, P., & Rosenfeld, L. (2007). Information Architecture for the World Wide Web. Sebastopol, CA: O’Reilly Media.