For example, the query needs some knowledge from some experts or specialist, like " why is little baby always crying", "how to make little baby of age 3-6 more brave", or I simply type " baby health" which I mean I want to search for knowledge about topics around "baby health". I tried these kinds of stuff but the result it returns is not what I want or I am not quite satisfied. I find that these questions or queries are not like "the longest river in US" or "the highest mountain", what it returns should be a kind of paragraphs which about some information on specific topic. Can anyone tell me if Wolfram Alpha can do this ? Thanks so much!
I think that the important thing to keep in mind when conceptualizing what Wolfram|Alpha is is that it addresses questions that are computational or data based. So, the things that the parser understands, generally speaking, are queries that imply some sort of computation or the retrieval of particular items of data, and then perhaps (usually) some computations based on those data. This might be information about a work--its definition, but also various statistical and linguistic information on that word, some of which is computed based on a combination of the data sources. One might argue that possible answers to the sorts of questions that you gave could fall into these categories of analysis--for example by returning relationships between various psychological and developmental theories, or outlining the historical progression of ideas in a field. But these sorts of things would require the curation of a very different set of classes of data (and many of which are either discursive or opinion oriented in many cases) than the sorts of explicitly factual data that are currently curated as inputs into the computations to answer computationally oriented queries.
Thanks a lot! So does Wolfram Alpha has such kind of data source that can answer " what are the stores that sell iphone 6", it seems that it is not computational based query, however, if Wolfram Alpha has stored such kind of data and then it may answer this query. The reason I am interested in such kind of stuff is that I think we can use the same technology that Wolfram Alpha uses to improve our search engineer, namely, to achieve our dream ---- one query, one answer (rather than a list of website which agrees on some keywords but are not related in terms of its content). And ever more, one query one answer that includes all aspects about the query, which is just what Wolfram Alpha does now! I am really interested in how Wolfram Alpha can achieve this and how it can understand human's language (query), as I understand Natural Language Processing has not achieved this ultimate goal by now. I wish that in the future after I type " what can I do if my baby is crying all the time" and then get the answer from a doctor or phychologist which is scientific.