Theses and Dissertations

Permanent URI for this collectionhttp://ir.daiict.ac.in/handle/123456789/1

Browse

Subject: search.filters.subject.Ontology

Search Results

Now showing 1 - 8 of 8
  • Thumbnail Image
    ItemOpen Access
    Simple automated story generator
    (Dhirubhai Ambani Institute of Information and Communication Technology, 2016) Khalpada, Purvish; Banerjee, Asim
    As technology advances, the list of things computers can do also increases. The list isreally big and includes diverse things. When we think about the list, storytelling is notsomething; we would be comfortable to have in the list. As we ask �why?� and try to come upwith an answer, we realize that human brain is a really wonderful thing. It can subconsciouslywork so hard and on an abstract level that while writing the stories, we don�t realize that thetask requires so much amount of world knowledge and involves many cognitive processes.The requirement of enormous data and complex processes makes this task difficult forcomputers. Difficult, but not impossible. This dissertation is one of the effort in that directionand the end product is SANDI (acronym for �Simple Automatis� Narrateur Des Intrigue�meaning �Simple Automated Story Generator�).SANDI can generate a story, either completely on its own or based upon the guidelineprovided by the user. For an example, a user may seed information about the charactersArjun and Aarohi and may provide an abstract storyline that Arjun and Aarohi meet andAarohi kill Arjun. Now, SANDI can create a story revolving around this abstract storyline,adding necessary events and twists, introducing new characters, if required. The final storyhas a flow, which is logical, coherent as well as interesting.
  • Thumbnail Image
    ItemOpen Access
    Ontology learning from relational database and reuse it with popular vocabularies
    (Dhirubhai Ambani Institute of Information and Communication Technology, 2015) Malodiya, Prakasha B.; Jat, P. M.
    Ontologies are the web documents generated by Web Ontology Languages that are used to develop semantic web. Semantic web requires data either in terms of manual creation or through conversion from existing data. A manual method for building ontologies is a very complex process and time-consuming. It also requires domain expert‘s need to understand the syntax and semantics of ontology development languages. It can also generate an error prone ontology. The data in the form of structured (relational) are used for ontology learning because they are most valuable data source available on the web. The research work for creating automatic ontology from structured database is not new. For this research work, many tools and methods were created to solve this type of problem. The primary limitation of the existing tools and methods for learning ontology from relational database is that the generated ontology is a simply copy of input database schema. This type of generated ontology gives the information from database schema and it does not contain any information about data. In this thesis we propose a tool for the automatic creation of ontology that gives the information about relational schema model and also the data stored in a database. Our aim is to analyze existing tools that were used for creating automatic ontology from relational databases and identify the advantages and disadvantages of these tools so that effective and valuable tool can be proposed. We have given detailed analysis of different existing tools used for creating automatic ontology from relational databases based on database schema and data stored in database and also performed a comparative analysis of these tools with our proposed tool.
  • Thumbnail Image
    ItemOpen Access
    Ontology alignment based support for searching LOD-datasets using SPARQL query
    (Dhirubhai Ambani Institute of Information and Communication Technology, 2015) Vijayvergiya, Nishant; Jat, P. M.
    The Linked Open Data (LOD) cloud is rapidly becoming the largest interconnected source of structured data on the web. Due to distributed nature of LOD and a growing number of ontologies and vocabularies used in LOD datasets,querying over multiple datasets and retrieving data from LOD-cloud remains a challenging task. The data on the LOD-cloud is stored in Resource Description Format(RDF) and SPARQL is the standard language for searching over RDF data. Different systems like LOQUS[1], ALOQUS[2] has been designed to formulate SPARQL queries which can obtain data from this LOD-cloud effectively and efficiently. Our work focusses on implementing a system like LOQUS,ALOQUS which uses SPARQL query to obtain data from the different LOD-datasets. But before querying on the datasets, mapping of the dataset specific ontologies with Upper Level Ontology is required, for formulating the SPARQL queries. In our work, we demonstrate ""Mapping Approach"" that is used for implementing our system, how two ontologies are mapped, how sparql queries are created using these mappings and how results are obtained after executing these queries.
  • Thumbnail Image
    ItemOpen Access
    Ontology development and query execution for an agro-advisory system
    (Dhirubhai Ambani Institute of Information and Communication Technology, 2014) Mordiya, Chetankumar; Bhise, Minal; Chaudhary, Sanjay
    In agriculture domain, farmers have queries regarding crop, soil, climate, cultivation process, disease, and pest. They express their queries in a natural language which are usually answered by agriculture experts. Due to lake of access, distance or time, the expert is usually not present physically to answer all the queries of the farmers. Hence, the farmers may not understand clearly what the experts wanted to convey. In such situation, there is a possibility of communication gap between farmers and knowledge of agriculture experts. It is desirable to capture agriculture experts’ knowledge in a system that understands farmers’ queries appropriately and gives the recommendations for it. An Agro-Advisory System is developed to fulfil these requirements. It is acknowledged based system. The knowledge base is maintained in the form of ontology. Ontology is integrated with services developed for this system. Ontology contains cotton crop knowledge of Gujarat region. Farmers can ask their queries related to cotton crop cultivation by Android device and get recommendations to improve crop productivity. The system is able to send notification and alert to farmers. Thesis work includes development of ontology for the cotton crop model and corresponding SPARQL (Simple Protocol and RDF Query Language) queries are executed on RDF (Resource Description Framework) data. Simple, complex and reasoning based queries are identified during thesis work.
  • Thumbnail Image
    ItemOpen Access
    Text normalization and non-monotonic knowledge base revision for consistent ontology learning
    (Dhirubhai Ambani Institute of Information and Communication Technology, 2013) Shah, Kushal; Dasgupta, Sourish
    0
  • Thumbnail Image
    ItemOpen Access
    SPARQL query optimization
    (Dhirubhai Ambani Institute of Information and Communication Technology, 2011) Singh, Rohit Kumar; Chaudhary, Sanjay
    Query Optimization is the process of selecting the most efficient query evaluation plan among the many strategies possible for processing a given query, especially if the query is complex. The users are not expected to write their queries in such a way so that they can be processed efficiently; rather it is expected from system to construct a query evaluation plan that minimizes the cost of query evaluation. In any query optimization, the goal is to find the execution plan which is expected to return the result set without actually executing the query or subparts with optimal cost. Query engines for ontological data mostly execute user queries without considering any optimization. Especially for large ontologies,optimization techniques are required to ensure that query results are delivered within reasonable time. SPARQL can be used to express queries across diverse data sources, whether the data is stored natively as RDF or viewed as RDF via middleware. So, Query optimization may speed up SPARQL query answering by knowledge intensive reformulation. In our research work, we have proposed learning approach to solve this problem. In our approach, the learning is triggered by user queries. Then the system uses an inductive learning algorithm to generate semantic rules. This inductive learning algorithm can automatically select useful join paths and properties to construct rules from a ontology with many concepts. The learned semantic rules are effective for optimization of SPARQL query because they match query patterns and reflect data regularities.
  • Thumbnail Image
    ItemOpen Access
    Context aware semantic service discovery
    (Dhirubhai Ambani Institute of Information and Communication Technology, 2009) Patel, Pankesh; Chaudhary, Sanjay
    This report is a part of an effort to develop a novel context based mechanism to facilitate semantic service discovery. Context- awareness is considered to be a key problem in designing more adaptive applications. Context modeling and reasoning are important research area of context-awareness computing. At the same time huge amount of data leading to inefficient systems raises a demand for filtering techniques. A study of existing discovery mechanisms has been done.
  • Thumbnail Image
    ItemOpen Access
    Design and development of an ontology for an agriculture information system
    (Dhirubhai Ambani Institute of Information and Communication Technology, 2007) Goyal, Shweta; Chaudhary, Sanjay
    Conceptual structures (Information theory); Knowledge representation (Information theory); Semantic networks (Information theory). Traditional sources of information like books, agricultural extension officer are unable to provide specific information required by a farmer. There is a need to build an ontology based agriculture information system which can provide scientific, relevant and contextual information about various aspects of crop production cycle. The aim is to develop a domain dependent ontology that will cover various aspects of crop production cycle. AGROVOC vocabulary developed by Food and Agriculture Organization is used for indexing and retrieving data in agricultural information systems. In the proposed research work, AGROVOC is used as base vocabulary to develop the proposed ontology (AGRIont). The ontology is developed by using open source tool Prot´eg´e 3.2. The ontology developed serves as a building block to an agriculture information system that answers the farmer’s queries in their own native language and helps them in making decisions about various aspects of crop production cycle. In this work, stages of ontology development, structure of ontology developed, architecture, query flow and process of query formulation for the proposed system are discussed in detail. At the end, guidelines to build an ontology, conclusions and future work are given.