Framework for Question-Answering in Sanskrit through Automated Construction of Knowledge Graphs

Sanskrit (sa\\d{m}sk\\d{r}ta) enjoys one of the largest and most varied\nliterature in the whole world. Extracting the knowledge from it, however, is a\nchallenging task due to multiple reasons including complexity of the language\nand paucity of standard natural language processing tools. In this paper, we\ntarget the problem of building knowledge graphs for particular types of\nrelationships from sa\\d{m}sk\\d{r}ta texts. We build a natural language\nquestion-answering system in sa\\d{m}sk\\d{r}ta that uses the knowledge graph to\nanswer factoid questions. We design a framework for the overall system and\nimplement two separate instances of the system on human relationships from\nmah\\=abh\\=arata and r\\=am\\=aya\\d{n}a, and one instance on synonymous\nrelationships from bh\\=avaprak\\=a\\'sa nigha\\d{n}\\d{t}u, a technical text from\n\\=ayurveda. We show that about 50% of the factoid questions can be answered\ncorrectly by the system. More importantly, we analyse the shortcomings of the\nsystem in detail for each step, and discuss the possible ways forward.\n