Search

Numerous sophisticated algorithms exist for discovering reoccurring patterns in financial time series. However, the most accurate techniques available produce opaque models, from which it is impossible to discern the rationale behind trading decisions. It is therefore desirable to sacrifice some degree of accuracy for transparency. One fairly recent evolutionary computational technology that creates transparent models, using a user-specified grammar, is grammatical evolution (GE). In this paper, we explore the possibility of evolving transparent entry- and exit trading strategies for the E-mini S&P 500 index futures market in a high-frequency trading environment using grammatical evolution. We compare the performance of models incorporating risk into their calculations with models that do not. Our empirical results suggest that profitable, risk-averse, transparent trading strategies for the E-mini S&P 500 can be obtained using grammatical evolution together with technical indicators. Best paper award.

IS researchers have so far developed conceptual propositions rather than empirical insights into what it takes to green an industry in practice. This paper analyzes an ongoing ten-year action design research effort, which seeks to leverage transport coopetition for environmental sustainability by guiding vertical standard design. Drawing on extensive field data, the paper theorizes about design challenges that surround attempts to tie together people, technologies, and processes into integrated IT solutions. In doing so, it illustrates the usefulness of a coopetitive lens for understanding these challenges and their effects on the evolution of such solutions. The paper also suggests that the action design research approach helps IS researchers to guide technology development in real-world situations.

Value creation of today is often a co-production in multi-organizational settings. This requires knowledge about how to conceive multi-organizational actor roles as foundations for co-ordinating and efficiently co-produce customer value. Some contemporary business process modelling approaches builds upon modelling interaction between two business parties (i.e. dyadic interaction), but do not acknowledge interaction patterns involving several network actors in their different actor roles. In this paper value creation in multi-organizational businesses are seen as value chains in value networks. The notion of assignments is the underlying structure in a multi-organizational perspective on business processes and is used to create foundations for distinguishing interaction patterns. Modelling and improving multi-organizational business processes conceived as action and interaction arranged in assignment structures, imply that dyadic role models need to be challenged as generative instruments. In this paper four generic multi-organizational network actor roles are brought forward (end- customer, main actor, co-ordinating actor, and co-producing actor) given meaning in and further instantiated in generic assignment actor roles based on their involvement in different multi-organizational interaction patterns. Thus, patterns of interaction constituting multi-organizational business processes are distinguished creating the necessary conditions for diverse network actors by the identification of their role in the action logic.

Within the field of information systems an interest in environmental issues has driven the agenda for research from green IT improvement to sustainable innovation. A challenge yet to investigate is how sustainable innovation involving a cluster of actors from multiple settings should be 1) designed and 2) orchestrated so that the innovation performed enables sustainable change. Processes for launching sustainable innovation should consequently be analysed in order to further investigate this notion. In northern Europe there is today a strong drive towards enabling initiatives utilizing mobile information technology improving the everyday transportation of people. This paper analysis the launch of a research and innovation cluster with the aim to develop information infrastructures and processes that stimulate distributed development of digital services for everyday travel. Events performed during the two-year start-up have been analysed identifying essential actions for network design and innovation orchestration, creating hypotheses, which enables further research about the establishment of sustainable innovation. Awarded Best Theme Paper Award

Information systems (IS) research on user involvement has primarily theorized relationships between developers, managers and users in systems development. However, so far, marginal attention has been paid to differences in user involvement practices between information systems. This paper explores user involvement in developing mobile and temporarily interconnected systems (MTIS). We refer to MTIS as heterogeneous systems that rely on network technologies for increasing the ubiquity of information services for users on the move. Such systems are becoming increasingly important in leveraging, e.g. car infotainment, supply chain management and wireless e-commerce. With particular emphasis on the nature of MTIS and its implications for user involvement, the paper analyses the systems development process of an action research project. The findings suggest that user involvement practices need to be adapted to accommodate features of this class of systems. Being an early attempt to trace the implications of technology features such as use context switches and temporary system relationships, the paper contributes to the development of an updated theory of the user role in an era of increased system complexity and stakeholder ambiguity.

In this paper the need for a multi-organizational perspective on business processes is outlined and a multi-grounded solution based on double-loop action research approach is proposed. Today, contemporary organizations’ capability to collaborate is an important competitive advantage and aligning in business networks is an increasingly common business model. Such development emphasizes the need for knowledge regarding how collaborative businesses could be characterized and how the constituent business interactions could be structured as several dyadic relationships in a multi-actor setting. The multi-organizational perspective proposed in this paper builds on pragmatic foundations and combines a language/action approach with a coordinative view on business processes, enabling design of complete action patterns. Haraldson, Sandra and Lind, Mikael. "The Emergence of a Multi-Organizational View on Business Processes – Experiences from a Double-loop Action Research Approach" (2010). AMCIS 2010 Proceedings. Paper 408.http://aisel.aisnet.org/amcis2010/408Sponsorship:University of Borås

This paper is challenging the usability of traditional usability guidelines. The claim is that guideline descriptions and explanations are not satisfactory. Analysis results demonstrate vagueness and are ambiguous in explanation. The aim of the paper is to propose a set of principles (meta-guidelines) to be used for improving the usability of guidelines.

In today’s business there is a clear need to find innovative procedures regarding product- or service development where several stakeholders meet in the same arena. An unresolved quest, however, is how such an arena could be set up and which activities to perform. This paper describes experiences from establishing such an arena, called an Internet Shopping Collaboratory (ISC). The ISC assembled researchers, practitioners, consumers, and solution providers in refining ideas to new products aimed to a future e-market. The basic idea has been to apply a co-design approach. The paper outlines why the ISC project did not work and lessons that were learnt. In the paper we make equivalent comparisons from the characteristics of the evolving ISC to the Living Lab concept. A focus on content was found vital for getting the different stakeholders engaged in the collaboratory. The case described uses an ideal scenario technique and applies a co-design approach.

This paper extends the conformal prediction framework to rule extraction, making it possible to extract interpretable models from opaque models in a setting where either the infidelity or the error rate is bounded by a predefined significance level. Experimental results on 27 publicly available data sets show that all three setups evaluated produced valid and rather efficient conformal predictors. The implication is that augmenting rule extraction with conformal prediction allows extraction of models where test set errors or test sets infidelities are guaranteed to be lower than a chosen acceptable level. Clearly this is beneficial for both typical rule extraction scenarios, i.e., either when the purpose is to explain an existing opaque model, or when it is to build a predictive model that must be interpretable. Sponsorship:This work was supported by the Swedish Foundation for Strategic Research throughthe project High-Performance Data Mining for Drug Effect Detection (IIS11-0053)and the Knowledge Foundation through the project Big Data Analytics by OnlineEnsemble Learning (20120192).

Online predictive modeling of streaming data is a key task for big data analytics. In this paper, a novel approach for efficient online learning of regression trees is proposed, which continuously updates, rather than retrains, the tree as more labeled data become available. A conformal predictor outputs prediction sets instead of point predictions; which for regression translates into prediction intervals. The key property of a conformal predictor is that it is always valid, i.e., the error rate, on novel data, is bounded by a preset significance level. Here, we suggest applying Mondrian conformal prediction on top of the resulting models, in order to obtain regression trees where not only the tree, but also each and every rule, corresponding to a path from the root node to a leaf, is valid. Using Mondrian conformal prediction, it becomes possible to analyze and explore the different rules separately, knowing that their accuracy, in the long run, will not be below the preset significance level. An empirical investigation, using 17 publicly available data sets, confirms that the resulting rules are independently valid, but also shows that the prediction intervals are smaller, on average, than when only the global model is required to be valid. All-in-all, the suggested method provides a data miner or a decision maker with highly informative predictive models of streaming data. Sponsorship:This work was supported by the Swedish Foundation for StrategicResearch through the project High-Performance Data Mining for Drug EffectDetection (IIS11-0053), the Swedish Retail and Wholesale DevelopmentCouncil through the project Innovative Business Intelligence Tools (2013:5)and the Knowledge Foundation through the project Big Data Analytics byOnline Ensemble Learning (20120192).QC 20180209

There has been little theorizing so far about the creation of new standards-based information services in public organizations. In this paper, we explore through a longitudinal case study at the Swedish Road Administration (SRA) how two standards - Alert-C and Location Code - were adapted as to deliver a traffic information service called RDS-TMC. Our in situ analysis reveals that the inherited norms, roles, and rules of the public organization hampered service delivery, which eventually created a tension between the old identity and the new identity of SRA - a tension we refer to as identity tension. Undergoing identity change, SRA had to deliberately configure infrastructural capabilities to better align its operational logics to the new service requirements. The findings suggest that digital multi-contextual services pose intriguing challenges for organizational identity among participating organizations.

In conformal prediction, predictive models output sets of predictions with a bound on the error rate. In classification, this translates to that the probability of excluding the correct class is lower than a predefined significance level, in the long run. Since the error rate is guaranteed, the most important criterion for conformal predictors is efficiency. Efficient conformal predictors minimize the number of elements in the output prediction sets, thus producing more informative predictions. This paper presents one of the first comprehensive studies where evolutionary algorithms are used to build conformal predictors. More specifically, decision trees evolved using genetic programming are evaluated as conformal predictors. In the experiments, the evolved trees are compared to decision trees induced using standard machine learning techniques on 33 publicly available benchmark data sets, with regard to predictive performance and efficiency. The results show that the evolved trees are generally more accurate, and the corresponding conformal predictors more efficient, than their induced counterparts. One important result is that the probability estimates of decision trees when used as conformal predictors should be smoothed, here using the Laplace correction. Finally, using the more discriminating Brier score instead of accuracy as the optimization criterion produced the most efficient conformal predictions. Sponsorship:Swedish Foundationfor Strategic Research through the project High-PerformanceData Mining for Drug Effect Detection (IIS11-0053) and theKnowledge Foundation through the project Big Data Analyticsby Online Ensemble Learning (20120192).

This paper explores the possibility of using the Hierarchical Temporal Memory (HTM) machine learning technology to create a profitable software agent for trading financial markets. Technical indicators, derived from intraday tick data for the E-mini S&P 500 futures market (ES), were used as features vectors to the HTM models. All models were configured as binary classifiers, using a simple buy-and-hold trading strategy, and followed a supervised training scheme. The data set was divided into a training set, a validation set and three test sets; bearish, bullish and horizontal. The best performing model on the validation set was tested on the three test sets. Artificial Neural Networks (ANNs) were subjected to the same data sets in order to benchmark HTM performance. The results suggest that the HTM technology can be used together with a feature vector of technical indicators to create a profitable trading algorithm for financial markets. Results also suggest that HTM performance is, at the very least, comparable to commonly applied neural network models.

In this paper we are presenting some theoreticalbackground, some practical applications and some futurescenarios of the use of the human being as a metaphor for designand implementation of e-services/avatars.The main conclusion is that e-services/avatars technology is apowerful concept but without a new profession as knowledgemanager in the background, there’s a big risk for failure. We arealso presenting a co-design model as a tool for the knowledgemanager.

Most predictive modeling techniques utilize all available data to build global models. This is despite the wellknown fact that for many problems, the targeted relationship varies greatly over the input space, thus suggesting that localized models may improve predictive performance. In this paper, we suggest and evaluate a technique inducing one predictive model for each test instance, using only neighboring instances. In the experimentation, several different variations of the suggested algorithm producing localized decision trees and neural network models are evaluated on 30 UCI data sets. The main result is that the suggested approach generally yields better predictive performance than global models built using all available training data. As a matter of fact, all techniques producing J48 trees obtained significantly higher accuracy and AUC, compared to the global J48 model. For RBF network models, with their inherent ability to use localized information, the suggested approach was only successful with regard to accuracy, while global RBF models had a better ranking ability, as seen by their generally higher AUCs.

Random forest is an often used ensemble technique, renowned for its high predictive performance. Random forests models are, however, due to their sheer complexity inherently opaque, making human interpretation and analysis impossible. This paper presents a method of approximating the random forest with just one decision tree. The approach uses oracle coaching, a recently suggested technique where a weaker but transparent model is generated using combinations of regular training data and test data initially labeled by a strong classifier, called the oracle. In this study, the random forest plays the part of the oracle, while the transparent models are decision trees generated by either the standard tree inducer J48, or by evolving genetic programs. Evaluation on 30 data sets from the UCI repository shows that oracle coaching significantly improves both accuracy and area under ROC curve, compared to using training data only. As a matter of fact, resulting single tree models are as accurate as the random forest, on the specific test instances. Most importantly, this is not achieved by inducing or evolving huge trees having perfect fidelity; a large majority of all trees are instead rather compact and clearly comprehensible. The experiments also show that the evolution outperformed J48, with regard to accuracy, but that this came at the expense of slightly larger trees. Sponsorship:This work was supported by the INFUSIS project www.his.se/infusis at the University of Skövde, Sweden, in partnership with the Swedish Knowledge Foundation under grant 2008/0502.

There has been a growing interest in the philosophy and constituents of design research by a vast amount of IS-scholars. There are several unre- solved concerns and issues in design research (DR). Some examples are the outcomes of design research, the role of theorizing in DR, how to conduct eval- uation and validation, and the need for different grounding processes to gener- ate valid knowledge from design research endeavors. This paper describes a multi-grounded approach for design research; consisting of three types of grounding processes (theoretical, empirical and internal grounding). The pur- pose is to investigate DR-based design knowledge and its roles during design research and design practice. A key feature in this approach is the division be- tween the meta-design (within design research) producing abstract design knowledge and the empirical design practice producing situational knowledge and artefacts. The multi-grounding approach to design research will be illus- trated by the support of two design cases. Sponsorship:University of Borås

Decision trees are often used for decision support since they are fast to train, easy to understand and deterministic; i.e., always create identical trees from the same training data. This property is, however, only inherent in the actual decision tree algorithm, nondeterministic techniques such as genetic programming could very well produce different trees with similar accuracy and complexity for each execution. Clearly, if more than one solution exists, it would be misleading to present a single tree to a decision maker. On the other hand, too many alternatives could not be handled manually, and would only lead to confusion. Hence, we argue for a method aimed at generating a suitable number of alternative decision trees with comparable accuracy and complexity. When too many alternative trees exist, they are grouped and representative accurate solutions are selected from each group. Using domain knowledge, a decision maker could then select a single best tree and, if required, be presented with a small set of similar solutions, in order to further improve his decisions. In this paper, a method for generating alternative decision trees is suggested and evaluated. All in all,four different techniques for selecting accurate representative trees from groups of similar solutions are presented. Experiments on 19 UCI data sets show that it often exist dozens of alternative trees, and that one of the evaluated techniques clearly outperforms all others for selecting accurate and representative models.

In this paper, data mining techniques are used toanalyze data gathered from online poker. The study focuses onshort-handed Texas Hold’em, and the data sets used containthousands of human players, each having played more than1000 hands. The study has two, complementary, goals. First,building predictive models capable of categorizing players intogood and bad players, i.e., winners and losers. Second,producing clear and accurate descriptions of what constitutesthe difference between winning and losing in poker. In theexperimentation, neural network ensembles are shown to bevery accurate when categorizing player profiles into winnersand losers. Furthermore, decision trees and decision lists usedto acquire concept descriptions are shown to be quitecomprehensible, and still fairly accurate. Finally, an analysis ofobtained concept descriptions discovered several ratherunexpected rules, indicating that the suggested approach ispotentially valuable for the poker domain.

When performing predictive data mining, the use of ensembles is known to increase prediction accuracy, compared to single models. To obtain this higher accuracy, ensembles should be built from base classifiers that are both accurate and diverse. The question of how to balance these two properties in order to maximize ensemble accuracy is, however, far from solved and many different techniques for obtaining ensemble diversity exist. One such technique is bagging, where implicit diversity is introduced by training base classifiers on different subsets of available data instances, thus resulting in less accurate, but diverse base classifiers. In this paper, genetic programming is used as an alternative method to obtain implicit diversity in ensembles by evolving accurate, but different base classifiers in the form of decision trees, thus exploiting the inherent inconsistency of genetic programming. The experiments show that the GP approach outperforms standard bagging of decision trees, obtaining significantly higher ensemble accuracy over 25 UCI datasets. This superior performance stems from base classifiers having both higher average accuracy and more diversity. Implicitly introducing diversity using GP thus works very well, since evolved base classifiers tend to be highly accurate and diverse.

Both theory and a wealth of empirical studies have established that ensembles are more accurate than single predictive models. Unfortunately, the problem of how to maximize ensemble accuracy is, especially for classification, far from solved. This paper presents a novel technique, where genetic algorithms are used for combining several measurements into a complex criterion that is optimized separately for each dataset. The experimental results show that when using the generated combined optimization criteria to rank candidate ensembles, a higher test set accuracy for the top ranked ensemble was achieved compared to using other measures alone, e.g., estimated ensemble accuracy or the diversity measure difficulty. Sponsorship:This work was supported by the Information Fusion Research Program (www.infofusion.se) at the University of Skövde, Sweden, in partnership with the Swedish Knowledge Foundation under grant 2003/0104.QC 20180209

This paper presents work in progress from theINFUSIS project and contains initial experimentation, usingpublicly available medicinal chemistry datasets, on obtainingcomprehensible QSAR models. Three techniques are evaluatedon both predictive performance, measured as accuracy, andcomprehensibility, measured as model size. The chosentechniques are J48 decision trees and JRip and Chipper decisionlists. The results show that J48 obtains superior accuracy andthat Chipper performs best of the two decision list algorithms onaccuracy. Furthermore, it is seen that, regarding accuracy, alltechniques benefit from feature reduction, which almost alwaysresults in increased accuracy. Regarding comprehensibility, JRipobtains the smallest models, followed by Chipper, with J48producing the largest models. For model size, feature reduction isnot seen to be universally beneficial; only J48 produces smallermodels for the reduced datasets, while both decision listalgorithms actually produce larger models on average. Theoverall conclusion is that, for these datasets, there exists a definitetradeoff between accuracy and comprehensibility that needs to beinvestigated further.

Genetic programming (GP), is a very general and efficient technique, often capable of outperforming more specialized techniques on a variety of tasks. In this paper, we suggest a straightforward novel algorithm for post-processing of GP classification trees. The algorithm iteratively, one node at a time, searches for possible modifications that would result in higher accuracy. More specifically, the algorithm for each split evaluates every possible constant value and chooses the best. With this design, the post-processing algorithm can only increase training accuracy, never decrease it. In this study, we apply the suggested algorithm to GP trees, extracted from neural network ensembles. Experimentation, using 22 UCI datasets, shows that the post-processing results in higher test set accuracies on a large majority of datasets. As a matter of fact, for two setups of three evaluated, the increase in accuracy is statistically significant. Sponsorship:This work was supported by the Information Fusion Research Program (University of Skövde, Sweden) in partnership with the Swedish Knowledge Foundation under grant 2003/0104 (URL: http://www.infofusion.se).

In this paper, several demands placed on concept description algorithms are identified and discussed. The most important criterion is the ability to produce compact rule sets that, in a natural and accurate way, describe the most important relationships in the underlying domain. An algorithm based on the identified criteria is presented and evaluated. The algorithm, named Chipper, produces decision lists, where each rule covers a maximum number of remaining instances while meeting requested accuracy requirements. In the experiments, Chipper is evaluated on nine UCI data sets. The main result is that Chipper produces compact and understandable rule sets, clearly fulfilling the overall goal of concept description. In the experiments, Chipper's accuracy is similar to standard decision tree and rule induction algorithms, while rule sets have superior comprehensibility.

The aim of the paper is to discuss the concept of communication quality in the context of the business process. The basic idea behind the concept of communication quality is that high quality communication id equal to a meaningful use of language. This implies that the concept off communication quality will be grounded in the philosophy of language and foremost in speech-act theory and is focused on the intersubjective and social aspects of quality in the business process.