Web Mining Research Articles

A Novel Mixed Wide and PSO-Bi-LSTM-CNN Model for the Effective Web Services Classification

In software technology, over the diversified environment, services can be rendered using an innovative mechanism of a novel paradigm called web services. In a business environment, rapid changes and requirements from various customers can be adapted using this service. For service management and discovery, the classification of Web services having the same functions is an efficient technique. However, there will be short lengthened Web services functional description documents, having less information, and sparse features. This makes difficulties in modelling short text in various topic models and leads to make an effect in the classification of Web services. A Mixed Wide and PSO-Bi-LSTM-CNN model (MW-PSO-Bi-LSTM-CNN) is proposed in this work for solving this issue. In this technique, the Web service category‟s breadth prediction is performed by combining Web services description document‟s discrete features, which exploits the wide learning model. In the next stage, the PSO-Bi-LSTM-CNN model is used for mining Web services description document word‟s context information and word order, for performing the Web service category‟s depth prediction. Here, particle swarm optimization (PSO) is integrated with the Bi-LSTM-CNN network for computing various hyper-parameters in an automatic manner. In third stage, Web service categories, results of depth, and breadth prediction are integrated using a linear regression model as final service classification result. At last, MW-PSO-Bi-LSTM-CNN, Wide&Bi-LSTM, and Wide&Deep web service classification techniques are compared and a better result with respect to web service classification accuracy is obtained using the proposed technique as shown in experimental results.

An Improved Rotation-Based Privacy Preserving Classification in Web Mining Using Naïve Bayes Classifier

An Improved Rotation-Based Privacy Preserving Classification in Web Mining Using Naïve Bayes Classifier

An Improved Rotation-Based Privacy Preserving Classification in Web Mining Using Naïve Bayes Classifier

An Improved Rotation-Based Privacy Preserving Classification in Web Mining Using Naïve Bayes Classifier

Incorporation of contextual information through Graph Modeling in Web content mining

Objectives: The objectives of this research article is to deal with the problem of web document clustering by modeling the web documents as directed completely labeled graphs that incorporate contextual information in the computation process to the extent required. The computational complexity of the MCS algorithm based on this graph model is O(n2), n being the number of nodes. As graph similarity using MCS is an NP-complete problem, so this is an important result that allows us to forgo sub-optimal approximation approaches and find the exact solution in polynomial time. Method: The first step towards this new approach of web document clustering is the representation of the web documents with the help of a directed completely labeled graph that can retain contextual information of the document under consideration. After graphical modeling of the document, the next step is the calculation of similarity between the graphical objects. For this purpose, a customized algorithm proposed as Algorithm for Maximum Common Subgraph Isomorphism (AMCSI)(1) based on a backtracking search scheme is being used. The proposed AMCSI algorithm is solving the problem of maximum common subgraph isomorphism in polynomial time. After obtaining the value for the similarity between the graphical objects we are again using a customized fuzzy-c means algorithm to produce clusters from the target set of web documents. We are using multidimensional scaling to express the distance values between the web pages (graphs) in two coordinates (x,y) and deterministic sampling to calculate the graph median in the process of fuzzy c-means clustering. Findings: We present an alternative method for web document clustering by representing the web documents as directed completely labeled graphs where the computational complexity of the MCS algorithm is O(n2)(1). A new distance measure is also developed based on the directed completely labeledgraph representation which is giving 16.9% better result than the prevailing methods(2). For the clustering purpose, we have chosen the fuzzy cmeans clustering algorithm and customizing the original algorithm to fit with graphical objects. This approach enables us to model the web documents as graphs without discarding contextual information and then cluster these graphical objects with the help of a well-established clustering algorithm. Keywords: Vector; graph; web document; subgraph; isomorphism; fuzzy; clustering

Automated User Behavior Mapping Using Web Usage Mining

Automated User Behavior Mapping is an application of web usage mining using which we can see the real-time behavior of end user visiting a particular web page automatically. The technologies used in this are socket programming for real-time communication between the server and the user accessing the website for collection of web log data and selenium web driver for automating the user behavior using web log files.

An Art of Web Services using Web Mining and Semantic Web Techniques

Todays web browsers serve as an easy access to numerous sources of text and multimedia data. More than a billion pages are indexed by search engines, and finding the desired information is not an easiest task. Over the last decade, there is an explosive growth in the information available on the World Wide Web (WWW). The objective of this paper is to provide an outline of web mining, its various classifications, its subtasks, and to give a perspective to the research community about the potential of applying techniques to extract meaningful patterns. This paper also gives information in the area of web services, semantic web mining and comparison of traditional web applications and semantic web applications thereby providing the guidelines for future research in the area of web services using web mining and semantic web.

A competitive optimization approach for data clustering and orthogonal non-negative matrix factorization

Partitioning a given data-set into subsets based on similarity among the data is called clustering. Clustering is a major task in data mining and machine learning having many applications such as text retrieval, pattern recognition, and web mining. Here, we briefly review some clustering related problems (k-means, normalized k-cut, orthogonal non-negative matrix factorization, ONMF, and isoperimetry) and describe their connections. We formulate the relaxed mean version of the isoperimetry problem as an optimization problem with non-negative orthogonal constraints. We first make use of a gradient-based optimization algorithm to solve this kind of a problem, and then apply a post-processing technique to extract a solution of the clustering problem. Also, we propose a simplified approach to improve upon solution of the 2-dimensional clustering problem, using the N-nearest neighbor graph. Inspired by this technique, we apply a multilevel method for clustering a given data-set to reduce the size of the problem by grouping a number of similar vertices. The number is determined based on two values, namely, the maximum and the average of the edge weights of the vertices connected to a selected vertex. In addition, using the connections between ONMF and k-means and between k-means and the isoperimetry problem, we propose an algorithm to solve the ONMF problem. A comparative performance analysis of our approach with other related methods shows outperformance of our approach, in terms of the obtained misclassification error rate and Rand index, on both benchmark and randomly generated problems as well as hard synthetic data-sets.

Public Security Sentiment Analysis on Social Web

This article presents (1) the results of a literature review on social web mining and sentiment analysis on public security; (2) the idea of a framework for the analytical process involved in the literature review themes; and (3) a research agenda with a perspective for future studies, considering some elements of the analytical process. The literature review was based on searches of five databases: Scopus, IEEE Xplore, Web of Science, ScienceDirect, and Springer Link. Search strings were applied to retrieve literature material of four kinds, without defining an initial time milestone, to get the historical register of publications associated with the main thematic. After some filtering, primary and secondary findings were separated, enabling the identification of elements for the framework. Finally, the research agenda is presented, containing a set of three research artifacts related to the proposed framework.

HPM: A Hybrid Model for User’s Behavior Prediction Based on N-Gram Parsing and Access Logs

The continuous growth of the World Wide Web has led to the problem of long access delays. To reduce this delay, prefetching techniques have been used to predict the users’ browsing behavior to fetch the web pages before the user explicitly demands that web page. To make near accurate predictions for users’ search behavior is a complex task faced by researchers for many years. For this, various web mining techniques have been used. However, it is observed that either of the methods has its own set of drawbacks. In this paper, a novel approach has been proposed to make a hybrid prediction model that integrates usage mining and content mining techniques to tackle the individual challenges of both these approaches. The proposed method uses N-gram parsing along with the click count of the queries to capture more contextual information as an effort to improve the prediction of web pages. Evaluation of the proposed hybrid approach has been done by using AOL search logs, which shows a 26% increase in precision of prediction and a 10% increase in hit ratio on average as compared to other mining techniques.

Performance Comparison of Data Mining Classifiers on Web Log Data

Web usage mining is the branch of web mining that deals with mining of data over the web. Web mining can be categorized as web content mining, web structure mining, web usage mining. In this paper, we have summarized the web usage mining results executed over the user tool WMOT (web mining optimized tool) based on the WEKA tool that has been used to apply various classification algorithms such as Naïve Bayes, KNN, SVM and tree based algorithms. Authors summarized the results of classification algorithms on WMOT tool and compared the results on the basis of classified instances and identify the algorithms that gives better instances accuracy.

Exploring the Potentialities of Automatic Extraction of University Webometric Information

Abstract Purpose The main objective of this work is to show the potentialities of recently developed approaches for automatic knowledge extraction directly from the universities’ websites. The information automatically extracted can be potentially updated with a frequency higher than once per year, and be safe from manipulations or misinterpretations. Moreover, this approach allows us flexibility in collecting indicators about the efficiency of universities’ websites and their effectiveness in disseminating key contents. These new indicators can complement traditional indicators of scientific research (e.g. number of articles and number of citations) and teaching (e.g. number of students and graduates) by introducing further dimensions to allow new insights for “profiling” the analyzed universities. Design/methodology/approach Webometrics relies on web mining methods and techniques to perform quantitative analyses of the web. This study implements an advanced application of the webometric approach, exploiting all the three categories of web mining: web content mining; web structure mining; web usage mining. The information to compute our indicators has been extracted from the universities’ websites by using web scraping and text mining techniques. The scraped information has been stored in a NoSQL DB according to a semi-structured form to allow for retrieving information efficiently by text mining techniques. This provides increased flexibility in the design of new indicators, opening the door to new types of analyses. Some data have also been collected by means of batch interrogations of search engines (Bing, www.bing.com) or from a leading provider of Web analytics (SimilarWeb, http://www.similarweb.com). The information extracted from the Web has been combined with the University structural information taken from the European Tertiary Education Register (https://eter.joanneum.at/#/home), a database collecting information on Higher Education Institutions (HEIs) at European level. All the above was used to perform a clusterization of 79 Italian universities based on structural and digital indicators. Findings The main findings of this study concern the evaluation of the potential in digitalization of universities, in particular by presenting techniques for the automatic extraction of information from the web to build indicators of quality and impact of universities’ websites. These indicators can complement traditional indicators and can be used to identify groups of universities with common features using clustering techniques working with the above indicators. Research limitations The results reported in this study refers to Italian universities only, but the approach could be extended to other university systems abroad. Practical implications The approach proposed in this study and its illustration on Italian universities show the usefulness of recently introduced automatic data extraction and web scraping approaches and its practical relevance for characterizing and profiling the activities of universities on the basis of their websites. The approach could be applied to other university systems. Originality/value This work applies for the first time to university websites some recently introduced techniques for automatic knowledge extraction based on web scraping, optical character recognition and nontrivial text mining operations (Bruni & Bianchi, 2020).

Semantic Network

Knowledge representation is an emerging field of research in Artificial Intelligence, Big data analytics, Semantic web, and Data Mining. Knowledge represented in an effective way helps in easy traversal, searching, reasoning, prediction, and inference. There are number of approaches, algorithms, techniques, and models that have been proposed for the same. Every approach has their own pros and cons. Hence our aim is to propose a simple and extremely effective way to represent knowledge, which has a greater expressiveness compared to logic and reverberates within the methods of people process data. In this paper we address the implementation of semantics network with simple yet powerful method, which despite the vague nature of English language, proves to give accurate results. The patterns discovered in sentences are listed in a defined order throughout the paper. The usage of Natural Language Toolkit and Posing tagging simplifies the most crucial task of tagging a word with its part of speech.

Web mining for innovation ecosystem mapping: a framework and a large-scale pilot study

Existing approaches to model innovation ecosystems have been mostly restricted to qualitative and small-scale levels or, when relying on traditional innovation indicators such as patents and questionnaire-based survey, suffered from a lack of timeliness, granularity, and coverage. Websites of firms are a particularly interesting data source for innovation research, as they are used for publishing information about potentially innovative products, services, and cooperation with other firms. Analyzing the textual and relational content on these websites and extracting innovation-related information from them has the potential to provide researchers and policy-makers with a cost-effective way to survey millions of businesses and gain insights into their innovation activity, their cooperation, and applied technologies. For this purpose, we propose a web mining framework for consistent and reproducible mapping of innovation ecosystems. In a large-scale pilot study we use a database with 2.4 million German firms to test our framework and explore firm websites as a data source. Thereby we put particular emphasis on the investigation of a potential bias when surveying innovation systems through firm websites if only certain firm types can be surveyed using our proposed approach. We find that the availability of a websites and the characteristics of the website (number of subpages and hyperlinks, text volume, language used) differs according to firm size, age, location, and sector. We also find that patenting firms will be overrepresented in web mining studies. Web mining as a survey method also has to cope with extremely large and hyper-connected outlier websites and the fact that low broadband availability appears to prevent some firms from operating their own website and thus excludes them from web mining analysis. We then apply the proposed framework to map an exemplary innovation ecosystem of Berlin-based firms that are engaged in artificial intelligence. Finally, we outline several approaches how to transfer firm website content into valuable innovation indicators.

Semantic Analysis to Identify Students’ Feedback

Abstract In this research, an automated analysis is performed on students’ chat and text data generated by social media platforms over the course of one semester and thoroughly analyzed for potential feedback about teaching, exams, and course contents. A data crawler is developed that performs horizontal and vertical samplings of the data. After data crawling, a few preprocessing steps are performed including text extraction, noise removal, stop-word removal, word stemming, text classification, and feature extraction. The intensity of a review is determined using four measures containing knowledge and understanding, course contents, teaching style, and assessment procedures for a specific course. The proposed system contains features from text mining and web mining to automatically identify a review whenever a user writes comments on their studies. This system aims to provide curriculum development committees with valuable online student feedback and assist in curriculum improvements. By comparing these automated reviews to results obtained from manual student survey forms, we found that the automated system yields the same output but at a fraction of the cost and time typically spent on collecting and analyzing manual student surveys.

WITHDRAWN: Experimental performance analysis of web recommendation model in web usage mining using KNN page ranking classification approach

Abstract Web service is the technique of detecting suitable service as per requirements and it’s applicable in enterprises, industry and government. In the real time web applications, the web usage mining is mainly focused on determining the knowledge about users and market trends. The purpose of this research is to explore the efficient web service model which plays crucial role for acquiring different services. Initially, an input data is taken from the dataset which consist of different web pages. Then the pre-processing step is included for accessing the web contents by users. For this purpose, the process of tag based pre-processing is performed. Afterwards, the modification method is performed where the user can alter, change or modify the contents in the web pages. From this research, the classification method in web usage mining can be utilized to find visitor activity information in web server logs which helps the organization for getting very useful support information and to specify the evaluation of web applications. In this paper, we introduced a new version of the web page recommendation model in web mining based on KNN page ranking classification algorithm. The recommendation model proposed is validated with social media dataset and the obtained results are tested with necessary parameters. The experimental results show our proposed model offers better performance over the existing algorithms with the high accuracy rate.

Understanding users' accessing behaviors to local Open Government Data via transaction log analysis

AbstractThe rapid development of Open Government Data (OGD) and the increasing attention on data use/reuse have stimulated many studies on data‐related issues. However, the findability of OGD is still one of the major challenges. Aiming to ameliorate the situation that “data is hard to find”, this paper examines OGD users' needs and accessing behaviors when interacting with local OGD portals. Transaction log analysis and web content mining were used in order to obtain insights from large groups of OGD users in an unobtrusive manner. Through analyzing transaction log data from three local OGD portals, including Open Data Philly (opendataphilly.org), Western Pennsylvania Regional Data Center (wprdc.org) and Analyze Boston (data.boston.gov), our study shows that users relied on different channels to enter local OGD portals, and such channels have different impacts on user success in finding the sought‐after data. We also find that OGD users prefer browsing over searching when inside the portals, the utilization of different browsing entries, and users' data needs.

Role of Educational Data Mining in Student Learning Processes With Sentiment Analysis

Educational data mining is a research field that is used to enhance education system. Research studies using educational data mining are in increase because of the knowledge acquired for decision making to enhance the education process by the information retrieved by machine learning processes. Sentiment analysis is one of the most involved research fields of data mining in natural language processing, web mining, and text mining. It plays a vital role in many areas such as management sciences and social sciences, including education. In education, investigating students' opinions, emotions using techniques of sentiment analysis can understand the students' feelings that students experience in academic, personal, and societal environments. This investigation with sentiment analysis helps the academicians and other stakeholders to understand their motive on education is online. This article intends to explore different theories on education, students' learning process, and to study different approaches of sentiment analysis academics.

Analysis of Health Care Services using Meta Search Engine and Web Mining

Analysis of Health Care Services using Meta Search Engine and Web Mining

Trajectory Based Location Prediction and Enriched Ontological User Profiles for Efficient Website Recommendation

The spread over of huge amount of information in the vast area of internet makes difficult for the users to obtain the search items that are relevant to them. The adoption of web usage mining helps to discover the accurate search results that satisfy their requirements. To fulfill their need, it is necessary to know their preferences of search at various contexts. In general, the user profiles are used to determine the taste of the users. The traditional method of user profiling does not provide a complete detail regarding their search. In addition, the search preference of the individuals varies in accordance with time and location. The user profiles do not update the dynamic location changes of the users. The traditional location based recommendation systems suggest the search results based on their location to compensate the dynamic preferences of the users. The drawbacks of the conventional systems are resolved by the Location and User Profile (LUP) based recommendation system. To attain a higher user satisfaction by providing accurate search results, a trajectory based location prediction and enriched ontological user profiles to recommend the appropriate websites to the users is proposed in this paper. In this article, we suggest a novel method for predicting the location of a user’s profile using Semantic Trajectory Pattern (STP), based on both the place and semantic features of user trajectories. Our prediction model ‘s central concept is based on a novel cluster-based prediction approach that evaluates the location of user search data based on the regular activities of related users in the same cluster, calculated by evaluating the typical behavior of users in semantic trajectories. The combination of location information along with enriched ontological user profiles improves the efficiency of the proposed web recommendation system. The experimental results are evaluated using recall, precision and F-measure metrics.

Monitoring and root-cause diagnostics of high-dimensional data streams

The high-dimensionality and volume of large-scale streaming data has inhibited significant research progress in developing an integrated monitoring and diagnostics (M&D) approach. Such data streams are becoming common in various applications including manufacturing, healthcare, and web mining. In this article, we propose an integrated M&D approach for large-scale streaming data. Using principal component analysis (PCA), we first develop a new monitoring method that adaptively chooses principal components that are most likely to be affected by the process change. Furthermore, we propose a novel diagnostic approach, seamlessly integrated with the proposed monitoring method to enable a streamlined SPC. This diagnostics approach draws inspiration from compressed sensing and uses adaptive lasso for identifying the sparse sources of the process change. We theoretically motivate our method and evaluate our integrated M&D method through simulations and case studies.