Beyond text-based web search: Question answering and by-example structured queries compared on Wikipedia

We must go beyond the limitations of text searching, and two user-friendly approaches have been proposed: the first is Question Answering (Q/A) and the second is By-Example Structured Queries (BEStQ) used by the SWiPE system. We compare the two approaches by defining an experimental setting based on the natural language questions made available by the QALD-4 challenge. In particular, we compare the results obtained with the BEStQ approach of SWiPE against those of Xser, a state-of-the-art Question Answering system, and against those obtained with the plain keyword search provided by the Wikipedia Search Engine. The experiments show that SWiPE outperforms the results provided by Wikipedia, and it also performs sensibly better than Xser, obtaining an overall 85% of totally correct answers vs. 68% of Xser. Among all answered questions, we obtain a precision of 100% and recall 96%. SWiPE is also able to answer more questions than the other systems. Then we outline simple ideas in which the two approaches can be integrated to maximize the power and the users' satisfaction.