Informatics Publishing Limited

K. Karpagam ¹, R. Balasubramanian ²

Affiliations
1 H.H. The Rajah's College (Autonomous), Pudukkottai, Tamil Nadu, IN
2 Dept. of Computer Applications, Karpaga Vinayaga College of Engineering and Technology, Kancheepuram, Tamil Nadu, IN

Abstract

Data mining aims at bringing out the hidden information from a large data set using data mining techniques according to the requirements. Association rule mining identifies itemsets that occur frequently in data set and frames association rules by taking all items equally. But many differences exist among the items that play a vital role in decision making. By taking one or more values of items as utilities, the utility mining technique works on finding the itemsets with greater utilities. In the proposed paper we present a utility mining algorithm named IUM (Improved Utility Mining) algorithm that finds high utility itemsets and also low utility itemsets from a large data set and the experiments states that the proposed algorithm performs better than existing algorithms in case of running time.

Keywords

Association Rules, Frequent Itemsets, Low Utility Itemset, High Utility Itemset.

Full Text

     

K. Karpagam ¹, A. Saradha ²

Affiliations
1 Department of Computer Applications, Dr. Mahalingam College of Engineering and Technology, IN
2 Department of Computer Science and Engineering, Institute of Road and Transport Technology, IN

Abstract

Human Computer Interaction is one of the pervasive application areas of computer science to develop with multimodal interaction for information sharings. The conversation agent acts as the major core area for developing interfaces between a system and user with applied AI for proper responses. In this paper, the interactive system plays a vital role in improving knowledge in the domain of health through the intelligent interface between machine and human with text and speech. The primary aim is to enrich the knowledge and help the user in the domain of health using conversation agent to offer immediate response with human companion feel.

Keywords

Artificial Intelligence, Question Answering, Conversational Agent, HCI, Pattern Matching, Speech Synthesis.

Full Text

     

R. Malathi ¹, A. Cholarajan ², K. Karpagam ¹, K. R. Jaya ¹, P. Muthukumaran ³

Affiliations
1 Department of Biotechnology, Bharathidasan University College (w), Orathanadu, Thanjavur, Tamilnadu, IN
2 Department of Microbiology, Meenakshi Chandrasekaran College of Arts and Science, Pattukkottai – 614 626, Thanjavur, Tamilnadu, IN
3 Department of Biochemistry, Meenakshi Chandrasekaran College of Arts and Science, Pattukkottai – 614 626, Thanjavur, Tamilnadu, IN

Abstract

Medicinal plants contribute in human health care system. Most of the plants utilizes by village people as a folk medicine. Now we are turned in to medicinal plant analysis of active compounds and conservation aspect. In the present study we had select the three important medicinal plants. We have collected three medicinally important medicinal plants such as Coleus amboinicus, Phyla nodiflora and Vitex negundo for antimicrobial studies. The experiment carried out in the selected medicinal plants leaves. The results are discussed with the available literature.

Keywords

Antimicrobial Activity, Pathogens, Coleus amboinicus, Phyla Nodiflora and Vitex Negundo.

Full Text

     

K. Karpagam ¹, A. Saradha ²

Affiliations
1 Department of Master of Computer Applications, Dr. Mahalingam College of Engineering and Technology, IN
2 Department of Computer Science and Engineering, Institute of Road and Transport Technology, IN

Abstract

Today, the information is growing enormously and it is difficult and tedious task to retrieve the necessary information from that pool. The main area for retrieving relevant answers is called intelligent information retrieval. To achieve this, question and answering system is used. This question and answering plays a major role in user query processing, information retrieval and extracting related information from the information pool. Recently, number of optimization algorithms is introduced to obtain the accurate and better results. Genetic Algorithm and Cuckoo Search are nature inspired meta-heuristic optimization algorithms. In this paper, combination of Genetic Algorithm with Cuckoo Search is applied to the question and answering system. The proposed algorithm is tested with the Amazon review, Trip Advisor and 20 news group data sets. The results are compared with Genetic Algorithm and Cuckoo Search algorithms.

Keywords

Document Clustering, Cuckoo Search, Genetic Algorithm, Information Retrieval, Question and Answering.

Full Text

     

K. Karpagam ¹, A. Saradha ²

Affiliations
1 Department of Master of Computer Application, Dr. Mahalingam College of Engineering & Technology, Pollachi, IN
2 Department of Computer science and Engineering, Institute of Road and Transport Technology, Erode., IN

Abstract

Document clustering or text clustering is an unsupervised technique and it is used to grouping the documents of same context. Document clustering algorithms are widely used in web searching engines to produce results relevant to a query. Today, the information in websites is growing in huge size and it leads to the process of managing, retrieve the required and updated information is a tedious task. Also necessary to obtain the exact information required by the user from the documents. Recently optimization algorithms are introduced and are applied to the clustering algorithms. The Genetic Algorithm and Cuckoo Search algorithms are meta-heuristic optimization algorithms and are used to obtain the optimum solutions. In this paper, Genetic Algorithm and Cuckoo Search algorithm based Domain-specific Keyword Similarity based Knowledgebase Creation algorithm are proposed to optimize the document clustering to answers the question answering system. The experimental were conducted on benchmark datasets and the performance was analyzed in terms of Precision, Recall, F1, Missrate, Fallout and Purity.

Keywords

Cuckoo Search, Document Clustering, Genetic Algorithm, Information Processing Knowledge Base, Text Mining.

Full Text

     

K. Karpagam ¹, K. Madusudanan ¹, A. Saradha ²

Affiliations
1 Department of Computer Applications, Dr. Mahalingam College of Engineering and Technology, IN
2 Department of Computer Science and Engineering, Institute of Road and Transport Technology, IN

Abstract

The conversation agent acts as core interfaces between a system and user in answering users queries with proper responses. Question answering system acquires an important role in the information retrieval field. The deep learning approach enhances the accuracy in answering complex questions. As outcome, the user is receiving the precise answer instead of large document collections. The aim of this paper is to develop a model with deep learning approach for improving answer selection process which supports more relevant answer displaying by conversation agents. To achieve this, word2vector used for word representation and biLSTM attentive model is used for training, testing and disclosure play precise answer. Question type is identified using POS-tagger based Question Pattern analysis (T-QPA) model. The knowledgebase is created from the bench mark datasets bAbI Facebook (simple QA tasks), TREC QA, Yahoo! Answer, Insurance QA dataset. The proposed framework is built by embedding of questions and answers based on bidirectional long short-term memory (biLSTM) attentive models. The similarity between questions and answers has been measured by semantic and cosine similarity. The proposed model reduces the search gap in extracting among user queries and answer sentences in the education domain. The system results are evaluated with the standard metrics MAP, Top 1 accuracy, F1- Score for the answer selection.

Keywords

Deep Learning, Question Answering System, Attentive Model, Conversation Agents, Cosine Similarity.

Full Text