Lab Assignment 7 - SaratM34/KDM-Lab-Assignments GitHub Wiki
Name: Mudunuri Sri Sai Sarat Chandra Varma
Class ID: 14
Objective: The Objective of this lab is to Generate a knowledge graph (ontology learning) as explained in tutorial in the following manner and Design your own ontology (classes, hierarchy, object and data properties) based on two dataset.
- Create a class hierarchy (classes) for a given domain (e.g., Food, Family) containing at least 5 classes: Created a class hierarchy with atleast 5 classes
- Create properties by specifying domains (subjects) and ranges (objects) (at least 5 properties): Created properties with specific domains and ranges
- Prepare a paragraph (in the same domain with your ontology) containing at least 5 sentences (containing at least 10 instances): Paragraph with sentences from created ontologies
Talks held on Gibraltar's future.Two days of talks on the future of Gibraltar begin at Jack Straw's country residence later on Wednesday.Officials at the two-day summit at the foreign secretary's official Kent house, Chevening, will plan a new forum on the Rock's future. In October, Mr Straw and his Spanish counterpart Miguel Moratinos agreed to establish a body that would give Gibraltarians a voice in their future. Most Gibraltarians said in a referendum they wanted to remain British. Gibraltar's Chief Minister Peter Caruana will represent the British citizens living on the Rock, while Britain's Europe Director Dominick Chilcott will represent the UK. Madrid is being represented by Spain's director general for Europe, Jose Maria Pons. The initiative follows Spain's socialist government's decision to put its long-standing sovereignty ambitions on hold. Gibraltarians rejected plans for the Rock's sovereignty to be shared between Britain and Spain in a referendum organised by Gibraltar government.
- Generate a knowledge graph including new Individuals and new Triplets by the execution of the knowledge graph generation program: Generated knowledge graph with new individuals and new triplets.
- Design at least 5 questions and execute the DL queries with your new knowledge graph:
i.Design your own ontology (classes, hierarchy, object and data properties) based on two datasets from the following datasets
a.Your own data
b.Stanford data https://rajpurkar.github.io/SQuAD-explorer/
c.Yahoo! Answer data (The training datacontains 2,698 questions, already labeled with one of the following 7 categories. The test datacontains 1,874 questions that are unlabeled).
- With Own Dataset
- With Yahoo Dataset
ii. Automatically create a set of individuals and triplets from your dataset based on classes and properties in your ontology: created a set of individuals and triplets from your dataset based on classes
iii. Formulate at least 10 valid and informative queries, implement them using DL queries, and execute them by mapping questions to DL queries. Report the results:
iv. Conduct Machine Learning Tasks with the Yahoo Questions/Answers datasets at https://umkc.box.com/s/4cvpv05cxets8jb2t5qf5vqvwoyrgxe3(at least 3 categories): Machine learning tasks with yahoo questions and answers