Compute query-based document summary

Improve an existing search service: help the user to understand the search results quicker and to formulate the search query more efficiently.
Compute query-based document summary
GetQueryBasedSummary
input: string Query, string[] Documents, enum DocumentFormat, int SummaryLength
output: Summary[] QueryBasedSummaryResult

Takes as an input a Query, a list of Documents and their DocumentFormat : text or URL. The document content is then summarized to a given SummaryLength, i.e. the number of core paragraphs to extract as a summary. Paragraphs, in which the Query appears are given preference.
Summary stores a collection of Paragraph.
Paragraph stores an Id and Score, and a collection of Sentence.
Sentence stores an Id, Score, Text and the Marked up sentence.

See also Document summarization.

Sample code in C#:

PingarAPIRequest request = new PingarAPIRequest();
request.AppID = "your app id";
request.AppKey = "your app key";
request.SearchSupport = new SearchSupportRequest();
request.SearchSupport.Query = "sample query";
request.SearchSupport.SummaryLength = 3;
request.SearchSupport.Documents = new string[] { "document text" };
request.SearchSupport.DocumentsFormat = DocumentFormat.Text;
request.Language = Language.EN;

PingarAPIServiceSoapClient pingarAPI = new PingarAPIServiceSoapClient();
PingarAPIResponse response = pingarAPI.GetQueryBasedSummary(request);
int count = 0;
if (response.Error == null)
{
    foreach (Summary summary in response.SearchSupport.QueryBasedSummaryResult)
    {
        Console.WriteLine("Summary for the document " + count);
        foreach (Paragraph paragraph in summary.Paragraphs)
        {
            foreach (Sentence sentence in paragraph.Sentences)
            {
                Console.WriteLine(sentence.Text);
            }
            Console.Write("\n\n");
        }
        count++;
    }
}

 
VIEW DEMO OF RAPID DISCOVERY COMPONENTS
 

Explore Pingar


Share Points CIO Apache Solr BizSpark