Prediction of User Future Request Utilizing the Combination of Both ANN and FCM in Web Page Recommendation

Raju, V.; Srinivasan, N.

doi:10.1515/jisys-2017-0310

Cited by 4 publications

(2 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Raju, V., & Srinivasan, N. [2] have explained the web page recommendation system, in the proposed system data has been categorized into potential and non-potential by use of an algorithm known as Levenberg-Marquardt firefly neural network algorithm, and the prediction is done by another clustering algorithm known as K-means algorithm. The recommendation accuracy is estimated corresponding to the various iterations.…”

Section: Literature Reviewmentioning

confidence: 99%

Classification of Web Pages Using the Machine Learning Algorithms with Web Page Recommendations

2022

IJIES

View full text Add to dashboard Cite

The World Wide Web holds a huge source of a variety of information. Web users are provided with a lot of information with numerous options and choices because of which decision-making will be difficult for the users. The users must be provided with a recommendation system that makes it easy for making their decisions, one such system is the web page recommendation system. The problem faced by web users is they can't get the required web pages when they search the web. To provide the solution to this problem web page recommendation systems are proposed.Recommendation system provides the users with interesting Webpages or websites where they can be reduced their searching time or surfing time. The web user needs effective and some suggestions for accessing the website efficiently. Therefore, the web recommendation systems are very useful for the user by efficiently handling the website. In the proposed work Web page recommendation system is implemented based on web page classification and by determining the page rank, where the classification of the web pages is done based on the machine learning algorithms. The classification techniques are used with the various machine learning algorithms in our work which are k-nearest neighbor, support vector machine, ADABOOSTER, and entropy-based ensemble Random Forest. The proposed algorithm in our research work is an entropy-based ensemble random forest. The novelty of this research work is the use of the entropy-based ensemble random forest. The web scraping technique has been used to fetch images and text from various websites. The data collected of 3000 are pre-processed and applied to the various classifiers to classify the web pages. The web pages were classified with the highest accuracy of 99.55% using the entropy-based ensemble random forest. The comparison of the entropy based random forest and the Gini-based random forest is shown to achieve the novelty. We have taken the data from the web pages in the form of images of around 17034 are classified them using the convolution neural network and pre-trained convolution neural network known as Resnet50. Compared its results with our proposed ensemble entropy-based random forest. Based on the web classification the web page recommendation is done. The page ranks for the collected web pages are calculated and the web page which will be similar to the given web page will be given as output. The page rank algorithm is used for the calculation of page rank. For every source website, a similar target web page is calculated. This is going to form the recommendation system.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Classification of Web Pages Using the Machine Learning Algorithms with Web Page Recommendations

2022

IJIES

View full text Add to dashboard Cite

show abstract

“…Clustering has been a rich subject of research in the area of recommendation systems and numerous clustering algorithms have been proposed in the literature [31]. This includes, but is not limited to, K-Means [30,32,33], Fuzzy C-Means [34][35][36], Bi-Clustering [37], Evolutionary Clustering [38][39][40], and Locality Sensitive Hashing [41].…”

Section: Introductionmentioning

confidence: 99%

OCA: Ordered Clustering-Based Algorithm for E-Commerce Recommendation System

et al. 2023

View full text Add to dashboard Cite

The industry of e-commerce (EC) has become more popular and creates tremendous business opportunities for many firms. Modern societies are gradually shifting towards convenient online shopping as a result of the emergence of EC. The rapid growth in the volume of the data puts users in a big challenge when purchasing products that best meet their preferences. The reason for this is that people will be overwhelmed with many similar products with different brands, prices, and ratings. Consequently, they will be unable to make the best decision about what to purchase. Various studies on recommendation systems have been reported in the literature, concentrating on the issues of cold-start and data sparsity, which are among the most common challenges in recommendation systems. This study attempts to examine a new clustering technique named the Ordered Clustering-based Algorithm (OCA), with the aim of reducing the impact of the cold-start and the data sparsity problems in EC recommendation systems. A comprehensive review of data clustering techniques has been conducted, to discuss and examine these data clustering techniques. The OCA attempts to exploit the collaborative filtering strategy for e-commerce recommendation systems to cluster users based on their similarities in preferences. Several experiments have been conducted over a real-world e-commerce data set to evaluate the efficiency and the effectiveness of the proposed solution. The results of the experiments confirmed that OCA outperforms the previous approaches, achieving higher percentages of Precision (P), Recall (R), and F-measure (F).

show abstract