A people-to-people matching system using graph mining techniques

A people-to-people matching system (or a match-making system) refers to a system in which users join with the objective of meeting other users with the common need. Some real-world examples of these systems are employer-employee (in job search networks), mentor-student (in university social networks), consume-to-consumer (in marketplaces) and male-female (in an online dating network). The network underlying in these systems consists of two groups of users, and the relationships between users need to be captured for developing an efficient match-making system. Most of the existing studies utilize information either about each of the users in isolation or their interaction separately, and develop recommender systems using the one form of information only. It is imperative to understand the linkages among the users in the network and use them in developing a match-making system. This study utilizes several social network analysis methods such as graph theory, small world phenomenon, centrality analysis, density analysis to gain insight into the entities and their relationships present in this network. This paper also proposes a new type of graph called “attributed bipartite graph”. By using these analyses and the proposed type of graph, an efficient hybrid recommender system is developed which generates recommendation for new users as well as shows improvement in accuracy over the baseline methods.

A social matching system : using implicit and explicit information for personalized recommendation in online dating service

Online dating websites enable a specific form of social networking and their efficiency can be increased by supporting proactive recommendations based on participants' preferences with the use of data mining. This research develops two-way recommendation methods for people-to-people recommendation for large online social networks such as online dating networks. This research discovers the characteristics of the online dating networks and utilises these characteristics in developing efficient people-to-people recommendation methods. Methods developed support improved recommendation accuracy, can handle data sparsity that often comes with large data sets and are scalable for handling online networks with a large number of users.

Pattern-based topic models for information filtering

Topic modelling, such as Latent Dirichlet Allocation (LDA), was proposed to generate statistical models to represent multiple topics in a collection of documents, which has been widely utilized in the fields of machine learning and information retrieval, etc. But its effectiveness in information filtering is rarely known. Patterns are always thought to be more representative than single terms for representing documents. In this paper, a novel information filtering model, Pattern-based Topic Model(PBTM) , is proposed to represent the text documents not only using the topic distributions at general level but also using semantic pattern representations at detailed specific level, both of which contribute to the accurate document representation and document relevance ranking. Extensive experiments are conducted to evaluate the effectiveness of PBTM by using the TREC data collection Reuters Corpus Volume 1. The results show that the proposed model achieves outstanding performance.

Process mining for clinical processes: A comparative analysis of four Australian hospitals

Business process analysis and process mining, particularly within the health care domain, remain under-utilised. Applied research that employs such techniques to routinely collected, health care data enables stakeholders to empirically investigate care as it is delivered by different health providers. However, cross-organisational mining and the comparative analysis of processes present a set of unique challenges in terms of ensuring population and activity comparability, visualising the mined models and interpreting the results. Without addressing these issues, health providers will find it difficult to use process mining insights, and the potential benefits of evidence-based process improvement within health will remain unrealised. In this paper, we present a brief introduction on the nature of health care processes; a review of the process mining in health literature; and a case study conducted to explore and learn how health care data, and cross-organisational comparisons with process mining techniques may be approached. The case study applies process mining techniques to administrative and clinical data for patients who present with chest pain symptoms at one of four public hospitals in South Australia. We demonstrate an approach that provides detailed insights into clinical (quality of patient health) and fiscal (hospital budget) pressures in health care practice. We conclude by discussing the key lessons learned from our experience in conducting business process analysis and process mining based on the data from four different hospitals.

Business process mining success

Process mining has developed into a popular research discipline and nowadays its associated techniques are widely applied in practice. What is currently ill-understood is how the success of a process mining project can be measured and what the antecedent factors of process mining success are. We consider an improved, grounded understanding of these aspects of value to better manage the effectiveness and efficiency of process mining projects in practice. As such, we advance a model, tailored to the characteristics of process mining projects, which identifies and relates success factors and measures. We draw inspiration from the literature from related fields for the construction of a theoretical, a priori model. That model has been validated and re-specified on the basis of a multiple case study, which involved four industrial process mining projects. The unique contribution of this paper is that it presents the first set of success factors and measures on the basis of an analysis of real process mining projects. The presented model can also serve as a basis for further extension and refinement using insights from additional analyses.

Pattern-based topics for document modelling in information filtering

Many mature term-based or pattern-based approaches have been used in the field of information filtering to generate users’ information needs from a collection of documents. A fundamental assumption for these approaches is that the documents in the collection are all about one topic. However, in reality users’ interests can be diverse and the documents in the collection often involve multiple topics. Topic modelling, such as Latent Dirichlet Allocation (LDA), was proposed to generate statistical models to represent multiple topics in a collection of documents, and this has been widely utilized in the fields of machine learning and information retrieval, etc. But its effectiveness in information filtering has not been so well explored. Patterns are always thought to be more discriminative than single terms for describing documents. However, the enormous amount of discovered patterns hinder them from being effectively and efficiently used in real applications, therefore, selection of the most discriminative and representative patterns from the huge amount of discovered patterns becomes crucial. To deal with the above mentioned limitations and problems, in this paper, a novel information filtering model, Maximum matched Pattern-based Topic Model (MPBTM), is proposed. The main distinctive features of the proposed model include: (1) user information needs are generated in terms of multiple topics; (2) each topic is represented by patterns; (3) patterns are generated from topic models and are organized in terms of their statistical and taxonomic features, and; (4) the most discriminative and representative patterns, called Maximum Matched Patterns, are proposed to estimate the document relevance to the user’s information needs in order to filter out irrelevant documents. Extensive experiments are conducted to evaluate the effectiveness of the proposed model by using the TREC data collection Reuters Corpus Volume 1. The results show that the proposed model significantly outperforms both state-of-the-art term-based models and pattern-based models

Oscar Pistorius and the death of Reeva Steenkamp : mapping public opinions on Twitter using frame analysis and data mining

This paper uses innovative content analysis techniques to map how the death of Oscar Pistorius' girlfriend, Reeva Steenkamp, was framed on Twitter conversations. Around 1.5 million posts from a two-week timeframe are analyzed with a combination of syntactic and semantic methods. This analysis is grounded in the frame analysis perspective and is different than sentiment analysis. Instead of looking for explicit evaluations, such as “he is guilty” or “he is innocent”, we showcase through the results how opinions can be identified by complex articulations of more implicit symbolic devices such as examples and metaphors repeatedly mentioned. Different frames are adopted by users as more information about the case is revealed: from a more episodic one, highly used in the very beginning, to more systemic approaches, highlighting the association of the event with urban violence, gun control issues, and violence against women. A detailed timeline of the discussions is provided.

Detecting news topics from microblogs using sequential pattern mining

This thesis presents a sequential pattern based model (PMM) to detect news topics from a popular microblogging platform, Twitter. PMM captures key topics and measures their importance using pattern properties and Twitter characteristics. This study shows that PMM outperforms traditional term-based models, and can potentially be implemented as a decision support system. The research contributes to news detection and addresses the challenging issue of extracting information from short and noisy text.

Mining universities : can we count on the integrity of research coming out of today's mining-funded tertiary sector?

Australia's economic growth and national identity have been widely celebrated as being founded on the nation's natural resources. With the golden era of pastoralism fading into the distance, a renewed love affair with primary industries has been much lauded, particularly by purveyors of neoliberal ideology. The considerable wealth generated by resource extraction has, despite its environmental and social record, proved seductive to the university sector. The mining industry is one of a number of industries and sectors (alongside pharmaceutical, chemical and biotechnological) that is increasingly courting Australian universities. These new public-private alliances are often viewed as the much-needed cash cow to bridge the public funding shortfall in the tertiary sector. However, this trend also raises profound questions about the capacity of public good institutions, as universities were once assumed to be, to maintain institutional independence and academic freedoms.

Facilitating query decomposition in query language modeling by association rule mining using multiple sliding windows

This paper presents a novel framework to further advance the recent trend of using query decomposition and high-order term relationships in query language modeling, which takes into account terms implicitly associated with different subsets of query terms. Existing approaches, most remarkably the language model based on the Information Flow method are however unable to capture multiple levels of associations and also suffer from a high computational overhead. In this paper, we propose to compute association rules from pseudo feedback documents that are segmented into variable length chunks via multiple sliding windows of different sizes. Extensive experiments have been conducted on various TREC collections and our approach significantly outperforms a baseline Query Likelihood language model, the Relevance Model and the Information Flow model.

Measuring patient flow variations : a cross-organisational process mining approach

Variations that exist in the treatment of patients (with similar symptoms) across different hospitals do substantially impact the quality and costs of healthcare. Consequently, it is important to understand the similarities and differences between the practices across different hospitals. This paper presents a case study on the application of process mining techniques to measure and quantify the differences in the treatment of patients presenting with chest pain symptoms across four South Australian hospitals. Our case study focuses on cross-organisational benchmarking of processes and their performance. Techniques such as clustering, process discovery, performance analysis, and scientific workflows were applied to facilitate such comparative analyses. Lessons learned in overcoming unique challenges in cross-organisational process mining, such as ensuring population comparability, data granularity comparability, and experimental repeatability are also presented.

Rabobank : a process mining case study BPI Challenge 2014 report

The increased interest in the area of process improvement persuaded Rabobank Group ICT in examining its own Change-process in order to improve its competitiveness. The group is looking for answers about the effectiveness of changes applied as part of this process, with particular interest toward the presence of predictive patterns and their parameters. We conducted an analysis of the log using well established process mining techniques (i.e. Fuzzy Miner). The results of the analysis conducted on the log of the process show that a visible impact is missing.

Mining business process deviance : a quest for accuracy

This paper evaluates the suitability of sequence classification techniques for analyzing deviant business process executions based on event logs. Deviant process executions are those that deviate in a negative or positive way with respect to normative or desirable outcomes, such as non-compliant executions or executions that undershoot or exceed performance targets. We evaluate a range of feature types and classification methods in terms of their ability to accurately discriminate between normal and deviant executions both when deviances are infrequent (unbalanced) and when deviances are as frequent as normal executions (balanced). We also analyze the ability of the discovered rules to explain potential causes and contributing factors of observed deviances. The evaluation results show that feature types extracted using pattern mining techniques only slightly outperform those based on individual activity frequency. The results also suggest that more complex feature types ought to be explored to achieve higher levels of accuracy.