Leveraging Individual and Collective Regularity to Profile and Segment User Locations from Mobile Phone Data

2021 | ACM Transactions on Management Information Systems | Citations: 0

Authors: Leng, Yan; Zhao, Jinhua; Koutsopoulos, Haris

Abstract: The dynamic monitoring of home and workplace distribution is a fundamental build ... Expand

Abstract: The dynamic monitoring of home and workplace distribution is a fundamental building block for improving location-based service systems in fast-developing cities worldwide. Inferring these places is challenging; existing approaches rely on labor-intensive and untimely survey data or ad hoc heuristic assignment rules based on the frequency of appearance at given locations. Motivated by the regularities in human behavior, we propose a novel method to infer the home, workplace, and third place based on an individual’s spatial-temporal patterns inferred from Call Detail Records. To capture the individual regularity, our method develops, for each person-location, the probability distribution that the person will appear in that location at a specific time of day using geo-temporal travel patterns a panel of individuals. To reveal the collective regularity, we apply eigen-decomposition to the matrix of historical geo-temporal data. Unsupervised machine learning techniques are then used to extract commonalities across locations for different groups of travelers, making inferences, such as home and workplace. Testing the methodology on real-world data with known location labels shows that our method identifies home and workplace with significant accuracy, improving upon the best practices in the literature by 79% and 34%, respectively. The methodology proposed is computationally efficient and is highly scalable to other real-world applications with historical tracking data. It provides a basis to improve location-based services, such as mobile commerce, social events recommendations, and urban transit design. Collapse

Topics: mobile phone wireless LAN global navigation satellite system mobile system information system use

Methods: cluster analysis computational algorithm k-means clustering principal component analysis data transformation

An Accurate and Scalable Role Mining Algorithm based on Graph Embedding and Unsupervised Feature Learning

2020 | HICSS | Citations: 0

Authors: Abolfathi, Masoumeh; Raghebi, Zohreh; Jafarian, Jafar Haadi; Banaei-Kashani, Farnoush

Abstract: Role-based access control (RBAC) is one of the most widely authorization models ... Expand

Abstract: Role-based access control (RBAC) is one of the most widely authorization models used by organizations. In RBAC, accesses are controlled based on the roles of users within the organization. The flexibility and usability of RBAC have encouraged organizations to migrate from traditional discretionary access control (DAC) models to RBAC. The most challenging step in this migration is role mining, which is the process of extracting meaningful roles from existing access control lists. Although various approaches have been proposed to address this NP-complete role mining problem in the literature, they either suffer from low scalability, or present heuristics that suffer from low accuracy. In this paper, we propose an accurate and scalable approach to the role mining problem. To this aim, we represent user-permission assignments as a bipartite graph where nodes are users and permissions, and edges are user-permission assignments. Next, we introduce an efficient deep learning algorithm based on random walk sampling to learn low-dimensional representations of the graph, such that permissions that are assigned to similar users are closer in this new space. Then, we use k-means and GMM clustering techniques to cluster permission nodes into roles. We show the effectiveness of our proposed approach by testing it on different datasets. Experimental results show that our approach performs accurate role mining, even for large datasets. Collapse

Topics: access control role-based access control

Methods: computational algorithm cluster analysis experiment fuzzy clustering optimization model

WHOSE MATURITY IS IT ANYWAY? THE INFLUENCE OF DIFFERENT QUANTITATIVE METHODS ON THE DESIGN AND ASSESSMENT OF MATURITY MODELS

2017 | European Conference On Information Systems | Citations: 0

Authors: Lasrado, Lester Allan; Vatrapu, Ravi; Mukkamala, Raghava Rao

Abstract: This paper presents results from an ongoing empirical study that seeks to unders ... Expand

Abstract: This paper presents results from an ongoing empirical study that seeks to understand the influence of different quantitative methods on the design and assessment of maturity models. Although there have been many academic publications on maturity models, there exists a significant lack of understanding of the potential impact of (a) choice of the quantitative approach, and (b) scale of measurement on the design and assessment of the maturity model. To address these two methodological issues, we analysed a social media maturity data set and computed maturity scores using different quantitative methods prescribed in literature. Specifically, we employed five methods (Additive, Variance, Cluster, Minimum Constraint, and RASCH) and compared the sensitivity of measurement scale and maturity stages. Based on our results, we propose a set of methodological recommendations for maturity model designers. Collapse

Topics: social media organizational value marketing management business process management

Methods: maturity model survey design fuzzy clustering cross sectional research design science

Theories: set theory

Possibilistic Clustering for Crisis Prediction: Systemic Risk States and Membership Degrees

2017 | HICSS | Citations: 0

Authors: Mezei, Jozsef; Sarlin, Peter

Abstract: Research on understanding and predicting systemic financial risk has been of incr ... Expand

Abstract: Research on understanding and predicting systemic financial risk has been of increasing importance in the recent years. A common approach is to build predictive models based on macro-financial vulnerability indicators to identify systemic risk at an early stage. In this article, we outline an approach for identifying different systemic risk states through possibilistic fuzzy clustering. Instead of directly using a supervised classification method, we aim at identifying coherent groups of vulnerability with macrofinancial indicators for pre-crisis data, and determine the level of risk for a new observation based on its similarity to the identified groups. The approach allows for differentiating among different possible pre-crisis states, and using this information for estimating the possibility of systemic risk. In this work, we compare different fuzzy clustering methods, as well as conduct an empirical exercise for European systemic banking crises. Collapse

Methods: cluster analysis machine learning experiment computational algorithm fuzzy clustering

Integrating expert knowledge and multilingual web crawling data in a lead qualification system

2016 | Decision Support Systems | Citations: 0

Authors: D’Haen, J.; Van den Poel, D.; Thorleuchter, D.; Benoit, D. F.

Abstract: Qualifying prospects as leads to contact is a complex exercise. Sales representa ... Expand

Abstract: Qualifying prospects as leads to contact is a complex exercise. Sales representatives often do not have the time or resources to rationally select the best leads to call. As a result, they rely on gut feeling and arbitrary rules to qualify leads. Model-based decision support systems make this process less subjective. Standard input for such an automated lead qualification system is commercial data. Commercial data, however, tends to be expensive and of ambiguous quality due to missing information. This study proposes web crawling data in combination with expert knowledge as an alternative. Web crawling data is freely available and of higher quality as it is generated by companies themselves. Potential customers use websites as a main information source, so companies benefit from correct and complete websites. Expert knowledge, on the other hand, augments web crawling data by inserting specific information. Web data consists of text that is converted to numbers using text mining techniques that make an abstraction of the text. A field experiment was conducted to test how a decision support system based on web crawling data and expert knowledge compares to a basic decision support system within an international energy retailer. Results verify the added value of the proposed approach. Collapse

Topics: website web crawling decision support system business-to-business database system

Methods: machine learning computational algorithm cluster analysis natural language processing experiment

Enriching iTunes App Store Categories via Topic Modeling

2014 | International Conference on Information Systems | Citations: 4

Authors: Vakulenko, Svitlana; Mã, Oliver

Abstract: Mobile application development is an emerging lucrative and fast growing market. ... Expand

Abstract: Mobile application development is an emerging lucrative and fast growing market. With the steady growth of the number of apps in the repositories the providers will inevitably face the need to fine-grain the existing hierarchy of categories used to organize the apps. In this paper we present a method to bootstrap the categorization process via topic modeling. We apply Latent Dirichlet Allocation (LDA) to the textual descriptions of iTunes apps in order to identify recurrent topics in the collection. We evaluate and discuss the results obtained from training the model on a set of almost 600,000 English-language app descriptions. Our results demonstrate that automated categorization via LDA-based topic modeling is a promising approach, that can help to structure, analyze and manage the content of app repositories. The topics produced complement the original iTunes categories, concretize and extend them by providing insights into the underlying category content. Collapse

Topics: mobile application mobile application market digital distribution Python user interface

Methods: topic model latent dirichlet allocation Gensim cluster analysis fuzzy clustering

Feedback reviews and bidding in online auctions: An integrated hedonic regression and fuzzy logic expert system approach

2013 | Decision Support Systems | Citations: 0

Authors: Zhang, Jie; Prater, Edmund L.; Lipkin, Ilya

Abstract: In online auctions, user-generated feedback reviews provide first-hand informati ... Expand

Abstract: In online auctions, user-generated feedback reviews provide first-hand information on the trustworthiness of transaction partners to the community. To examine how the feedback reviews are taken into account of the buyers' bidding decisions and thus affect the final winning price of an auction, we thoroughly examine how buyers mentally interpret the seller's reviews and adjust the bids accordingly. With ample bidding results data from a popular auction website eBay.com, this paper adopts an integrated approach of Fuzzy Logic Expert System (FLES) model and a statistical hedonic regression model to examine the research question. In particular, we use the hedonic regression approach to select key variables, which are then entered into a FLES analysis to generate knowledge base regarding the relationships between variables such as item characteristics, auction characteristics and review scores, and the final winning price. This integrated approach combines the advantages of both methods, and also overcomes their own limitations. In addition, we also present the insights gained from bidding behaviors utilizing each of the approaches. Collapse

Topics: knowledge base online auction website electronic commerce fuzzy logic

Methods: regression analysis method cluster analysis computational algorithm fuzzy clustering theory development

Theories: fuzzy set theory

Identifying generic routings for product families based on text mining and tree matching

2007 | Decision Support Systems | Citations: 3

Authors: Jiao, Jianxin (Roger); Zhang, Lianfeng (Linda); Pokharel, Shaligram; He, Zhen

Abstract: Product customization leads to an exponentially increased number of product and ... Expand

Abstract: Product customization leads to an exponentially increased number of product and process variants, which exaggerates the difficulties in building up customization capabilities for make-to-order production systems. It is imperative for companies to configure existing operations routings by exploiting similarities among product and process families so as to take advantage of repetitions. Corresponding to a product family, a process family comprises a set of similar production processes that share certain common operations routings (namely, generic routings). In addition to leveraging the costs of delivering variety, exploiting process families around generic routings can reduce development risks by reusing existing facilities and proven process elements. This paper applies data mining techniques to identify generic routings from large amount of production information and process data available in a firm's legacy systems. Generic routing identification encompasses three consecutive stages, including routing similarity measure, routing clustering and routing unification. Text mining and tree matching techniques are applied to cope with the textual and structural types of data underlying generic routings. A case study of mass customization production of vibration motors for mobile phones is reported to illustrate the feasibility and potential of generic routing identification. Collapse

Topics: mobile phone

Methods: machine learning qualitative content analysis case study cluster analysis natural language processing

Convergence or coalescence?Information technology and the reshaping of industry boundaries

2006 | European Conference On Information Systems | Citations: 1

Authors: Poon, Simon; Davis, Joseph; Peters, Georg

Abstract: The impacts of the evolving information technology (IT) on organisations, market ... Expand

Abstract: The impacts of the evolving information technology (IT) on organisations, markets, industries, and economies have been researched intensively. Aspects studied have included productivity, employment, business value, markets, organisation, and industry structure, to name a few. With the increasing availability of panel data spanning longer periods, it is now possible to study the relatively longerterm effects of IT and its role in industry coalescence and the reshaping of industry boundaries. Collapse

Topics: IT investment productivity IS impact mainframe internet technology

Methods: cluster analysis hierarchical linear modeling literature study fuzzy clustering computational algorithm

Theories: theory of economic cycles

FUZZY CLUSTERING AS AN ALTERNATIVE CLASSIFICATION METHOD IN INFORMATION SYSTEMS RESEARCH

2002 | Americas Conference on Information Systems | Citations: 0

Authors: Sethi, Vikram; Eakin, Mark; Barrier, Tonya; Duffy, Kevin; Sethi, Vijay

Abstract: Cluster Analysis is a powerful exploratory method that has been used in pattern ... Expand

Abstract: Cluster Analysis is a powerful exploratory method that has been used in pattern analysis, grouping, decisionmaking, and pattern classification. Cluster analysis is particularly appropriate for the exploration of interrelationships among the data points to make an assessment of their structure. General tasks in cluster analysis (Jain, Murty, and Flynn 1999) involve pattern representation, definition of pattern proximity, clustering or grouping, and assessment of out. Within the broader scope of information systems research, cluster analysis has been used in the area of end-user computing, assessment of IS strategies, global information systems research and assessment of supply chain configurations. In most cases, researchers have adopted hard or crisp cluster analysis whereby each respondent or analysis unit is assigned a class label, or becomes a part of one and only one cluster (unitary membership). In several cases, unitary membership may not reflect situation realities. An alternative clustering methodology – fuzzy clustering - provide an alternative analytical mechanism to unitary membership by allowing gradual membership in different groups, with membership values indicating the probability or degree of membership within a specific group or cluster. We illustrate fuzzy clustering with an example from the field of end-user computing and then develop general guidelines for the broader use of the method. Collapse

Topics: end user computing systems design user requirement systems implementation

Methods: computational algorithm cluster analysis fuzzy clustering survey survey design