Conference Papers
Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/123456789/28506
Browse
9 results
Search Results
Item Web user session clustering using modified K-means algorithm(2011) Poornalatha, G.; Raghavendra, P.S.The proliferation of internet along with the attractiveness of the web in recent years has made web mining as the research area of great magnitude. Web mining essentially has many advantages which makes this technology attractive to researchers. The analysis of web user's navigational pattern within a web site can provide useful information for applications like, server performance enhancements, restructuring a web site, direct marketing in e-commerce etc. The navigation paths may be explored based on some similarity criteria, in order to get the useful inference about the usage of web. The objective of this paper is to propose an effective clustering technique to group users' sessions by modifying K-means algorithm and suggest a method to compute the distance between sessions based on similarity of their web access path, which takes care of the issue of the user sessions that are of variable length. © 2011 Springer-Verlag.Item Clustering web page sessions using sequence alignment method(2011) Poornalatha, G.; Raghavendra, S.R.This paper illustrates clustering of web page sessions in order to identify the users' navigation pattern. In the approach presented here, user sessions of variable lengths are compared pair wise, numbers of alignments are found between them and the distances are measured. Web page sessions are clustered by employing the modified k-means algorithm. A couple of web access logs including the well known NASA data set are used to illustrate the effectiveness of the clustering. R-squared measure is applied to determine the optimal number of clusters and chi-squared test is carried out to see the association between the various web page sessions that are clustered. These two measures show the goodness of the clusters formed. © 2011 Springer-Verlag.Item Similarity analysis of service descriptions for efficient Web service discovery(Institute of Electrical and Electronics Engineers Inc., 2014) Kamath S․, S.; Ananthanarayana, V.S.Web services are currently one of the preferred ways for realizing Service Oriented Architectures in business systems. Due to this popularity and also due partly to the failure of the Universal Business Registry initiative, the number of published service descriptions openly available on the Web has increased by a large extent and hence, relevant service discovery as per user specification remains a challenge. In order to achieve more efficient discovery, we propose a crawler based system for gathering service descriptions available on the Web for building a scalable service repository. We apply similarity analysis techniques to the service descriptions after extracting features provided by the service descriptions and automatically generate relevant tags for each service. Using Agglomerative Hierarchical clustering, we cluster the tagged service descriptions and use the same tagging technique to generate tags for each cluster. For generating cluster tags, we take into account how well the tag represents the corresponding service in the cluster and how well the service itself represents the cluster it is in. The search domain for service discovery was significantly reduced by tagging & clustering and and we show that our system achieves good results. © 2014 IEEE.Item Recommendation of Optimal Locations for Government Funded Educational Institutes in Urban India Using a Hybrid Data Mining Technique(Institute of Electrical and Electronics Engineers Inc., 2015) Pulakhandam, S.; Patil, N.The Government of India has introduced schemes to build educational facilities in areas where literacy rate is less than the national average. It was found that literacy rate is a sufficient criterion with respect to rural areas but a different approach must be taken for urban planning because of space constraints, heterogeneous communities and the varied background of children living in urban areas. A hybrid data mining method to discover optimum locations for educational facilities in urban areas is proposed. The method is a combination of rule-based classification and spatial clustering. Rule-based classification is used to identify relevant data points from the spatial data set. New parameters like dropout rate and ratio of children out of school to children in school are introduced to measure relevance since literacy rate alone was found to be an insufficient criterion. Spatial clustering is used to group the points according to their location. The center of each cluster signifies the optimum location for an educational facility. A modified COD-CLARANS method is proposed. The algorithm is modified in two aspects. It is proposed that the absolute error, E, is calculated using the shortest path of commute on city roads rather than the obstructed distance calculated in the pre-processing step of the original COD-CLARANS algorithm. Secondly, only areas with space available for the establishment of a facility are considered to represent clusters. The modified method seeks improve efficiency and to make the spatial clustering technique more relevant to the urban setting. A comparison between different clustering algorithms and the modified COD-CLARANS algorithm is presented. © 2015 IEEE.Item ECABBO: Energy-efficient clustering algorithm based on Biogeography optimization for wireless sensor networks(Institute of Electrical and Electronics Engineers Inc., 2019) Nomosudro, P.; Mehra, J.; Naik, C.; Shetty D, D.Cluster-based communication design is an assuring technique in wireless sensor networks to reduce energy consumption and enhance scalability. The requirement of data collection from neighbor nodes, data gathering, and data forwarding to the sink overloads each cluster head. Therefore, it is a highly significant issue to elect a set of optimal cluster heads from the normal sensor nodes. In this paper, the Biogeography Based Optimization for energy-efficient clustering is introduced for cluster head selection. The simulation outcomes show that the algorithm improves the network endurance as compared to other protocols such as Genetic algorithm, Low energy adaptive clustering hierarchy, and Clustered Routing for Selfish Sensors. © 2019 IEEE.Item Effectiveness of Phase Correlation Spectral Similarity Measure in Distinguishing Target Signatures for Hyperspectral Data Analysis(Institute of Electrical and Electronics Engineers Inc., 2020) Yadav, P.P.; Shetty, A.; Raghavendra, B.S.; Narasimhadhan, A.V.Hyperspectral imaging is one of the most information-rich sources of remote sensing data that exists. It can capture the entire, continuous spectrum of color and light. Feature extraction techniques that are selected for identifying diagnostic features influence classification accuracy. Spectral matching algorithms like similarity measures are developed to compare spectral features of materials with their reference spec-tral signatures in identifying different earth surfaces. Similarity measures are used as simple feature extraction techniques in target identification using hyperspectral data. Though there are several similarity measures, selecting a robust similarity measure requires further investigation. Influence of similarity measures are not studied much in distinguishing spectrally distinct signatures. In this article, we propose to study the performance of similarity measures in (i) discriminating endmember signatures (ii) mixed pixel identification (iii) clustering signatures of different classes and (iv) endmember extraction. Experimental results show an effective and a robust performance of proposed phase correlation similarity measure among all other similarity measures compared for all the problems under investigation. © 2020 IEEE.Item Gradient Based Spectral Similarity Measure for Hyperspectral Image Analysis(Institute of Electrical and Electronics Engineers Inc., 2021) Yadav, P.P.; Shetty, A.; Raghavendra, B.S.; Narasimhadhan, A.V.Spectral matching algorithms or similarity measures dis-criminate spectral signatures of unknown materials by comparing them with the reference spectral signatures. Spectral matching algorithms play an important role in identifying different earth surface features using hyperspectral image analysis. Identification of diagnostic features and discriminating spectral signatures, especially of mineral signatures, which exhibit subtle differences among themselves is still a challenging task. Thus, developing or coming up with a new spectral matching algorithms is expected. Therefore, in this paper, we present a gradient based spectral similarity measure (GSSM) that captures the diagnostic (absorption) features to measure the degree of closeness between spectral signatures. Effectiveness of the proposed GSSM in distinguishing spectrally distinct signatures is studied with that of other spectral matching algorithms in (i) discriminating endmember signatures (ii) mixed pixel identification (iii) clustering spectral signatures of different classes and (iv) endmember extraction. Experimental results on a spectral library of 5 minerals formed using a benchmark mineral dataset called Cuprite data clearly show that the proposed GSSM is capable of (i) discriminating different spectral signatures in a better manner and also (ii) bringing into notice of the user the locations of diagnostic features by highlighting them to recognize easily. © 2021 IEEE.Item Gradient Correlation Incorporated Similarity Measures in Matching Spectral Signatures(Institute of Electrical and Electronics Engineers Inc., 2022) Yadav, P.P.; Shetty, A.; Raghavendra, B.S.; Narasimhadhan, A.V.Hyperspectral images provide ample information which is needed to be analyzed carefully by the spectral processing algorithms for identifying objects, finding minerals etc. Spectral matching algorithms (SMAs) which make use of similarity measures discriminate and identify earth surface features by comparing spectral signatures with the ground-truth. SMAs that discriminate overall patterns capturing the diagnostic features of spectral signatures is of great use. In view of this, in this paper, we explore spectral gradient as a diagnostic feature to discriminate spectral signatures. Applicability of the proposed spectral gradient which is incorporated with SMAs in distinguishing spectrally distinct signatures is experimented in the following cases: (i) discriminating endmember signatures (ii) mixed pixel identification (iii) clustering spectral signatures of different classes and (iv) endmember extraction. Overall, the experimental results on a benchmark mineral Cuprite dataset library of five minerals have shown significantly improved performance in discriminating various spectral signatures. © 2022 IEEE.Item Semi-Automatic Labeling and Semantic Segmentation of Gram-Stained Microscopic Images from DIBaS Dataset(Institute of Electrical and Electronics Engineers Inc., 2023) Chethan Reddy, G.P.; Reddy, P.A.; Kanabur, V.R.; Vijayasenan, D.; Sumam David, S.; Govindan, S.In this paper, a semi-Automatic annotation of bacteria genera and species from DIBaS dataset is implemented using clustering and thresholding algorithms. A Deep learning model is trained to achieve the semantic segmentation and classification of the bacteria species. Pixel-level classification accuracy of 95 percent is achieved. Deep learning models find tremendous applications in biomedical image processing. Automatic segmentation of bacteria from gram-stained microscopic images is essential to diagnose respiratory and urinary tract infections, detect cancer, etc. Deep learning will aid the biologists to get reliable results in less time. Additionally, a lot of human intervention can be reduced. This work can be helpful to detect bacteria from urinary smear images, sputum smear images, etc to diagnose urinary tract infections, tuberculosis, pneumonia, etc. © 2023 IEEE.