Log File Analysis

Nowadays, large transportation household surveys cannot be conducted without the help of powerful management and support tools, and the information technologies are useful for preparing, conducting, and post-analyzing such surveys. In the Greater Montreal Area (GMA), the 2003 household survey followed the general methodology that has been developed over the past twenty years to integrate the finest software, databases, and methods. The tools making up the household survey information system (HSIS) are based on the Totally Dissagregate Approach and its object-oriented extension. This paper presents the background and the fundamentals of the Montreal 2003 survey information system, and describes the way in which it has been assembled, illustrating the functional and technical architectures that were used. It also emphasizes the transposability of the method to other transportation survey activities and planning tools. The final discussion stresses the "winning" elements involved in conducting a modern transportation household survey successfully.

Simulation of human users engaged in interactive information retrieval (IIR) may be a key resource to enable evaluation of IR systems in interactive settings in ways that are scalable, objective, and cheap to implement. This paper considers generation of simulated users in an operationalist framework. It identifies several challenges due to the cognitive nature of IIR and its highly conditionalized interactions with information systems and suggests a program for user simulation must rely on results from user studies and will need to overcome several difficult modeling problems.

WebQuilt is a web logging and visualization system that helps web design teams run usability tests (both local and remote) and analyze the collected data. Logging is done through a proxy, overcoming many of the problems with server-side and client-side logging. Captured usage traces can be aggregated and visualized in a zooming interface that shows the web pages people viewed.

This paper describes the visual analysis tool WebQuilt, a web usability logging and visualization system that helps web design teams record and analyze usability tests. The logging portion of WebQuilt unobtrusively gathers clickstream data as users complete specified tasks. This data is then aggregated and presented as an interactive graph, where nodes of the graph are images of the web pages visited, and arrows are the transitions between pages. To aid analysis of the gathered usability test data, the WebQuilt visualization provides filtering capabilities and semantic zooming, allowing the designer to understand the test results at the gestalt view of the entire graph, and then drill down to sub-paths and single pages. The visualization highlights important usability issues, such as pages where users spent a lot of time, pages where users get off track during the task, navigation patterns, and exit pages, all within the context of a specific task. WebQuilt is designed to conduct remote usability testing on a variety of Internet-enabled devices and provide a way to identify potential usability problems when the tester cannot be present to observe and record user actions.

WebQuilt is a web logging and visualization system that helps web design teams run usability tests (both local and remote) and analyze the collected data. Logging is done through a proxy, overcoming many of the problems with server-side and client-side logging. Captured usage traces can be aggregated and visualized in a zooming interface that shows the web pages people viewed.

Understanding the effects of different functionality in complex learning environments is one of the key challenges of learning technology research. Particularly for new personal learning environments, which combine a variety of services to offer new educational approaches such comparisons become increasingly important for analysing the effects compared to established learning environments. This paper discusses the use of cross-system log file analysis for enabling the comparison between different logging approaches used either ...

Due to the rapid growth of the use of computers and Internet in education a large number of Web Based E-Learning (WBEL) systems have been developed and implemented. However, most these systems do not consider the learner’s motivation level during the learning process, while motivation is one of the most important factors for student achievement in learning. In this paper we argue that motivation assessment is very important for WBEL systems, therefore more research is needed in this area. A critical review of previous research on motivation assessment in e-Learning systems is also presented, and finally suggestions of how to assess the learner’s motivation during a learning activity in a WBEL system. Benefits from our approach would maintain a learner’s motivation during the learning process resulting in a higher performance level within the WBEL system.

This study sought to determine whether infusion device event logs could support accident investigation. Methods: An incident reporting database was searched for information about log file use in investigations. Log file data from devices in clinical use were downloaded and electronically searched for characteristics (signatures) matching specific function queries. Different programming sequences were simulated, and device logs were downloaded for analysis. Results: Database reports mentioned difficulties resolving log file data to the incident report and used log file data to confirm programming failures. Log file search revealed that, aside from alarm types and times, the devices were unable to adequately satisfy functional queries. Different simulated programming scenarios could not be easily differentiated by log file analysis. Conclusions: The device logs we studied collect data that are poorly suited to accident investigation. We conclude that infusion device logs cannot function as black boxes do in aviation accidents. Logs would be better applied to assist routine operations.

PATTERNS AT WORK: PREDEFINED PROPERTY SPECIFICATIONS FOR THE ANALYSIS OF DISTRIBUTED SYSTEMS The notion of patterns is increasingly being used in the design, test, and analysis of systems, e.g., in the Rational Suite (recently acquired by IBM) and the Bandera project (University of Kansas). In particular, patterns that encode system requirements and can be used in the analysis phase are getting more common in the field of software engineering. Here, we describe an approach that targets the problem of analyzing and testing distributed systems, and whose implementation-within the ObjectGEODE (OG) toolkit of Telelogic-relies on the concept of property specification patterns. The approach consists of using traces of distributed systems to regenerate a model of the system under test in the Specification and Description Language (SDL) and using the property specification language of ObjectGEODE, GOAL, to express properties of interest. The model checker of OG is then used to verify the system against the specified properties. To facilitate the task of developers/debuggers, we construct a library of parameterized property patterns in GOAL that can be used in the verification of various applications. We also discuss how the library of GOAL patterns and the possible manipulations on them can add to the usability of the language and the toolkit.

Component-based Software Engineering (CBSE) is concerned with the development of systems from reusable parts (components), and the development and maintenance of these parts. This study addresses the issue of usability testing in a CBSE environment, and specifically automatically measuring the usability of different components in a single system. The proposed usability measure is derived from the message exchange between components recorded in a log file. The measure was validated in an experimental evaluation. Four different prototypes of a mobile telephone were subjected to usability tests, in which 40 subjects participated. Results show that the usability of the individual components can be measured, and that they can be priorities on their potential for improvement.

Betingelser for brug af denne artikel Denne artikel er omfattet af ophavsretsloven, og der må citeres fra den. Følgende betingelser skal dog vaere opfyldt:  Citatet skal vaere i overensstemmelse med "god skik"  Der må kun citeres "i det omfang, som betinges af formålet"  Ophavsmanden til teksten skal krediteres, og kilden skal angives, jf. ovenstående bibliografiske oplysninger. Søgbarhed Artiklerne i de aeldre LexicoNordica (1-16) er skannet og OCR-behandlet. OCR står for 'optical character recognition' og kan ved tegngenkendelse konvertere et billede til tekst. Dermed kan man søge i teksten. Imidlertid kan der opstå fejl i tegngenkendelsen, og når man søger på fx navne, skal man vaere forberedt på at søgningen ikke er 100 % pålidelig.

Generating entities recommendations has attracted considerable interest in recent years. Most recently published works mainly focus on providing a user with the most relevant and/or personalized entity recommendations that score highly against the query and/or the user's preference. Some works consider user side information, such as the user network, user relations, and user's demographic information, and propose to integrate them into the framework of recommender systems. These approaches have been shown to increase the users' satisfaction and engagement with the system. In this paper, we investigate entities recommender systems and summarize the recent efforts in this domain by categorizing approaches. The first category presents different approaches that utilize knowledge graph as side information. The second category gathers work that consider both the current query, and the users' previous interactions with the system. These latter works have considered the full user history to personalize the ranking of recommended entities related to the query. In this review paper, we emphasize contextual information-based approaches that utilize user's context and feedback to improve the recommendations. We accomplished a summary of the literature and synthesized the papers according to different perceptions. Finally, a comparison between approaches is provided and some drawbacks are identified.

E-Learning is a rapid humanizing as a cost-effective education paradigm and providing the best ways to the institutions to expand their educational programs to provide easy access to a large set of student populations. Student motivation to learn electronically has a key concerning issue with e-learning system. Learners' performance progress and success based on e-learner's motivation level and type. It's very crucial to detect and improve the motivational aspect of an e-learning course at any level. With this research we focus on student's motivational approach and used student motivation as a basic parameter to model an intelligent e-learning supportive framework for the e-learners. In addition, additional modules are added to the standard intelligent tutoring framework system to make one to one communication i.e. of the learner and tutor. Mainly three strategies namely learning, motivation and tutoring further based on several key aspects have been incorporated in the proposed model, with the inclusion of two additional modules named as, student current level module and student motivation module in the basic ITS models. The modules of the proposed system are such as interface module, Student current level module, student motivation module, tutoring module, student performance module and expert module. Student module is termed as student performance module in the proposed framework, for maintaining the continuous track of learner's performance while keeping him/her engaged with the learning activity and the tutor.

Electronic tables running single display groupware have good potential for enhancing face-to-face collaborations. Our goal is to understand how people collaborate over physical and electronic tables. We articulate the unique collaborative characteristics of tables, and describe our initial observations of people collaborating on a regular table, a table display, and a vertical display.

This research focuses on determining whether a student’s grit impacts their behavior within an intelligent tutoring system, towards developing better student models and feature sets that can help tutors predict student behavior and determine whether tutors might foster improvements in students’ grit, perseverance and recovery from failure. We use rare Association Rule Mining to explore how students’ grit may be associated with students’ behavior within MathSpring, an intelligent tutoring system, as a first step.

This research focuses on determining whether a student’s grit impacts their behavior within an intelligent tutoring system, towards developing better student models and feature sets that can help tutors predict student behavior and determine whether tutors might foster improvements in students’ grit, perseverance and recovery from failure. We use rare Association Rule Mining to explore how students’ grit may be associated with students’ behavior within MathSpring, an intelligent tutoring system, as a first step.

Problem/Background Understanding the difference between correlation and causation is a critical concept in scientific explanation. The distinction is instantiated in the Next Generation Science Standards (NGSS) in the Cross-Cutting Concepts of "Patterns" and "Cause and Effect" (Achieve, 2013). While the standards suggest that "patterns can be used to identify cause and effect relationships" (PAT-M3) and that "patterns of change can be used to make predictions" (PAT-E2), the standards also recognize that "empirical evidence is required to differentiate between cause and correlation and make claims about specific causes and effects" (CE-H1). Helping students to realize that patterns can be used to identify possible cause and effect relationships and that scientist have means to intervene on patterns to discern causation-thus differentiating co-variation and causation-are important goals for science education. As science educators, it would help us to more deeply understand these distinctions and how students come to understand them.

The growing need for content customization in websites has fostered the development of systems which try to identify the user's navigation patterns. These may be, normally, identified by means of log file analysis. However, this solution does not identify the semantic intention behind user's navigation. This paper provides an approach to incorporating semantic knowledge to the process of identifying the user's intentions in the navigation of a website with semantic support. The capture of the user's intentions is achieved by the semantic enrichment of the log files and the use of and approach that takes into account the linguistic and cognitive aspects in the development of the user model.

In the process of knowledge discovery, workers examine available information in order to make sense of it. By sensemaking, we mean interacting with and operating on the information with a variety of information processing mechanisms (3,18). Previously, we introduced a concept that uses the spreadsheet metaphor with cells containing visualizations of complex data. In this paper, we extend and apply

Several visualizations have emerged which attempt to visualize all or part of the World Wide Web. Those visualizations, however, fail to present the dynamically changing ecology of users and documents on the Web. We present new techniques for Web Ecology and Evolution Visualization (WEEV). Disk Trees represent a discrete time slice of the Web ecology. A collection of Disk Trees forms a Time Tube, representing the evolution of the Web over longer periods of time. These visualizations are intended to aid authors and webmasters with the production and organization of content, assist Web surfers making sense of information, and help researchers understand the Web.

The information present on web is in different formats, to access those different crawling methods are available which gives more focus on either document or content. Information retrieved by these crawlers was relevant according to specific IP. Here we introduce a new approach for profile based crawling i.e. Profile Based Document Specific Crawling System which leads the user to the document he is searching for. This system is useful to collect information from social media sites, which is relevant according to document and more specific about user who has uploaded that. Further, divide a information into two parts, first semantic data, which comes from the user's own contribution, and user's social contacts. Second i.e. descriptive data, which gives information about document for which we are in need. Important part in this will be ranking of user profile depending on information uploaded without accessing actual information and also without taking users personal information.

Surfing the World Wide Web (WWW) involves traversing hyperlink connections among documents. The ability to predict surfing patterns could solve many problems facing producers and consumers of WWW content. We analyzed WWW server logs for a WWW site, collected over ten days, to compare different path reconstruction methods and to investigate how past surfing behavior predicts future surfing choices. Since log files do not explicitly contain user paths, various methods have evolved to reconstruct user paths. Session times, number of clicks per visit, and Levenshtein Distance analyses were performed to show the impact of various reconstruction methods. Different methods for measuring surfing patterns were also compared. Markov model approximations were used to model the probability of users choosing links conditional on past surfing paths. Information-theoretic (entropy) measurements suggest that information is gained by using longer paths to estimate the conditional probability of link choice given surf path. The improvements diminish, however, as one increases the length of path beyond one. Information-theoretic (total divergence to the average entropy) measurements suggest that the conditional probabilities of link choice given surf path are more stable over time for shorter paths than longer paths. Direct examination of the accuracy of the conditional probability models in predicting test data also suggests that shorter paths yield more stable models and can be estimated reliably with less data than longer paths.

Resumo: Neste artigo, apresentamos alguns aspectos concernentes a aproximações teóricas entre o pensamento computacional e a abordagem microgenética de Barbel Inhelder. O objetivo é investigar de que maneira a microgênese pode elucidar os processos cognitivos que estão envolvidos no desenvolvimento de habilidades relativas ao pensamento computacional. Para tanto, utilizamos como metodologia a pesquisa bibliográfica, explorando os principais conceitos presentes em estudos que versam sobre as duas temáticas. Os resultados apontam que o conhecimento dos elementos da microgênese podem fundamentar teoricamente as estratégias do pensamento computacional. Dessa forma, argumentamos que uma análise do desenvolvimento cognitivo, por meio da microgênese, poderá trazer contribuições importantes para apoiar propostas pedagógicas para o desenvolvimento do pensamento computacional. Palavras-chave: pensamento computacional; microgênese cognitiva; resolução de problemas. Computational thinking under the gaze of cognitive microgenesis Abstract: In this article, we present some aspects concerning theoretical approximations between computational thinking and Barbel Inhelder's microgenetic approach. The objective is to investigate how microgenesis can elucidate the cognitive processes that are involved in the development of skills related to computational thinking. For this, we used bibliographical research as methodology, exploring the main concepts present in studies that deal with the two themes. The results indicate that the knowledge of the elements of microgenesis can theoretically support the strategies of computational thinking. Thus, we argue that an analysis of cognitive development, through microgenesis, may bring important contributions to support pedagogical proposals for the development of computational thinking.

In this paper we describe a method for evaluating the usability of web-based applications. Our method is based on remote and automatic capture and semi-automatic analysis of users behavior, in order to find usability problems in the applications' interfaces. The goal of our method is to allow an analysis of the way users actually interact with the evaluated interface. Through the analysis of users behavior is possible to find patterns of interaction. Analyzing the patterns found and comparing it to the expected behavior for the tasks performed by users, we can detect usability problems. In this paper we also briefly describe a first experiment with our method and some initial results that point to the potential of the method in performing remote and automatic usability evaluations.

The Web has become a ubiquitous environment for application delivery. The originally intended idea, as a distributed system for knowledgeinterchange, has given way to organizations offering their products and services using the Web as a global point of sale. The centralized delivery-mechanism enables the construction of E-Commerce applications personalized for each user by using behavior analysis. Current technologies suffer from the Web's legacy and use Log file-analysis or collaborative filtering only to adapt the content to users' needs. Motivated by the results of collaborative filtering algorithms, we describe a construction approach based on the abstract concept of services. To support the fine-grained concept we use the component-based WebComposition Markup Language to support reuse and seamless evolution of E-Commerce applications.

In this tutorial, we introduce two rule-based systems for on and offline trace analysis, RULER and LOGSCOPE. RULER is a conditional rule-based system, which has a simple and easily implemented algorithm for effective runtime verification, and into which one can compile a wide range of temporal logics and other specification formalisms used for runtime verification. Specifications can be parameterized with data, or even with specifications, allowing for temporal logic combinators to be defined. We outline a number of simple syntactic extensions of core RULER that can lead to further conciseness of specification but still enabling easy and efficient implementation. RuleR is implemented in Java and we will demonstrate its ease of use in monitoring Java programs. LOGSCOPE is a derivation of RULER adding a simple very user-friendly temporal logic. It was developed in Python, specifically for supporting testing of spacecraft flight software for NASA's next 2011 Mars mission MSL (Mars Science Laboratory). The system has been applied by test engineers to analysis of log files generated by running the flight software. Detailed logging is already part of the system design approach, and hence there is no added instrumentation overhead caused by this approach. While post-mortem log analysis prevents the autonomous reaction to problems possible with traditional runtime verification, it provides a powerful tool for test automation. A new system is being developed that integrates features from both RULER and LOGSCOPE.

In this thesis a study of the use and users of the Swedish online, monolingual, learner's dictionary Lexins svenska lexikon, LSL, is presented. The target group of LSL is learners at a relatively ad vanced beginner's level. In 2008-2011, the LSL underwent a comprehensive revision and the study is based on data from before (LSL3) and after (LSL4) the revision. The thesis has a threefold aim, namely to 1) investigate the use of LSL, 2) find out who the users of LSL are, and 3) take a closer look at individual informant's search behaviour. This, in turn, forms the basis for three sub studies. A general question to be answered is whether the LSL is actually used by the dictionary's target group and if so, how? In sub study 1 the use of LSL is examined by means of a log file analysis which produces a general view of the users' most frequent look-ups. The data from before (i.e. 2008) and after (i.e. 2011) the revision are strikingly different. For example, the 2011 data contains more central and common Swedish words than the 2008 data. The users of LSL4 appear, to a greater extent, to belong to the target group of LSL, than the users of LSL3. In sub study 2 a web questionnaire survey is used to find out information about the users of LSL which results in an important, though generalized, user profile. Evidently, both LSL3 and LSL4 are used by the target group, but, to judge from the two sets of questionnaire data, to a greater extent in the updated version of LSL, i.e. LSL4. Lastly, in sub study 3 log file data and questionnaire data are triangulated which made it possible to observe more closely individual target group informants and their dictionary activities. This qualitative study results in yet another picture of the use of LSL. To sum up, combining methods and triangulating data seem to have resulted in a more complete and balanced picture of the use of LSL.

In the process of knowledge discovery, workers examine available information in order to make sense of it. By sensemaking, we mean interacting with and operating on the information with a variety of information processing mechanisms (3,18). Previously, we introduced a concept that uses the spreadsheet metaphor with cells containing visualizations of complex data. In this paper, we extend and apply

To ensure the normal operation of a large computer network system, the common practice is to constantly collect system logs and analyze the network activities for detecting anomalies. Most of the analysis methods in use today are highly automated due to the enormous size of the collected data. Conventional automated methods are largely based on statistical modeling, and some employ machine learning. In this paper, we show interactive visualization as an alternative and effective data exploration method for understanding the complex behaviors of computer network systems. We describe three log-file analysis applications, and demonstrate how the use of our visualization-centered tools can lead to the discovery of flaws and intruders in the network systems.

Bytecode instrumentation is a preferred technique for building profiling, debugging and monitoring tools targeting the Java Virtual Machine (JVM), yet is fundamentally dangerous. We illustrate its dangers with several examples gathered while building the DiSL instrumentation framework. We argue that no Java platform mechanism provides simultaneously adequate performance, reliability and expressiveness, but that this weakness is fixable. To elaborate, we contrast internal with external observation, and sketch some approaches and requirements for a hybrid mechanism.

This paper describes the visual analysis tool WebQuilt, a web usability logging and visualization system that helps web design teams record and analyze usability tests. The logging portion of WebQuilt unobtrusively gathers clickstream data as users complete specified tasks. This data is then aggregated and presented as an interactive graph, where nodes of the graph are images of the web pages visited, and arrows are the transitions between pages. To aid analysis of the gathered usability test data, the WebQuilt visualization provides filtering capabilities and semantic zooming, allowing the designer to understand the test results at the gestalt view of the entire graph, and then drill down to sub-paths and single pages. The visualization highlights important usability issues, such as pages where users spent a lot of time, pages where users get off track during the task, navigation patterns, and exit pages, all within the context of a specific task. WebQuilt is designed to conduct remote usability testing on a variety of Internet-enabled devices and provide a way to identify potential usability problems when the tester cannot be present to observe and record user actions.

Given the important role that meta-cognitive processes play in learning, intelligent tutoring systems should not only provide domain-specific assistance, but should also aim to help students in acquiring meta-cognitive skills. As a step toward this goal, we have constructed a Help Tutor, aimed at improving students' help-seeking skill. The Help Tutor is based on a cognitive model of students' desired help-seeking processes, as they work with a Cognitive Tutor (Aleven et al., 2004). To provide metacognitive tutoring in conjunction with cognitive tutoring, we designed an architecture in which the Help Tutor and a Cognitive Tutor function as independent agents, to facilitate re-use of the Help Tutor. Pilot tests with four students showed that students improved their help-seeking behavior significantly while working with the Help Tutor. The improvement could not be attributed to their becoming more familiar with the domain specific skills being taught by the tutor. Although stud...

UAIC: Faculty of Computer Science, “Alexandru Ioan Cuza” University, Romania {florin.cristea, vlad.alexa, adiftene}@info.uaic.ro ... Abstract. In this paper, we summarize our analysis of the large log of multilingual image searches in Flickr provided to iCLEF 2009 participants, focusing on the UAIC group. ... H.3 [Information Storage and Retrieval]: H.3.1 Content Analysis and Indexing; H.3.3 Information Search and Retrieval; H.4 [Information Systems Applications]: H.4.m Miscellaneous ... Interactive Information Retrieval, Cross-language ...

Discovering of behaviour patterns of website visitors is one of the most common applications in web log mining. Based on the discovered users' behaviour patterns, it is possible to restructure or in combination with other knowledge personalize the examined website, portal or other web-based system. Data preparation represents the first inevitable step in the process of discovering users' behavioural patterns. In this paper we summarize the results of our previous research, where we carefully examined the relevance of individual steps of data preparation from a web server log file and virtual learning environment for further analysis. The aim of our experiments was to find out to what extent it is necessary to realize the time-consuming data preparation in the process of discovering patterns of behaviour of web users and to determine the inevitable steps to obtain reliable data from different types of log files. Considering the obtained results we propose a methodology for data preparation in the process of discovering patterns of web user behaviour based on the results of experiments we carried out. The research results showed, that in the case of systems providing sophisticated navigation options and a rigid structure of the content (which is characteristic for the most virtual learning environments), the paths completing is not an inevitable step in data preparation in the process of discovering patterns of web users' behaviour.

While much has been written about the importance of innovation, there is still much to learn about the specific behaviours that lead to innovation among groups. In this paper we introduce a framework of innovation based on behaviours identified as being conducive to collaborative innovation. We also report on a study of a task designed to elicit innovation supportive behaviours in a virtual world environment. The task resulted in a variety of solutions and a range of participant behaviours, and specific behaviours were correlated with innovative solutions. Multiple forms of analysis provided unique insights into participant behaviour, and the combined set of analyses led to a richer understanding of participant behaviour than found through any individual analysis. The paper also presents implications for how organizations may scaffold group interactions to increase the chances of successful collaborative innovation.

Objectives: Automated piggybacks are purported to make drug administration safer and more reliable. We evaluated the human factors of piggyback infusion, investigated the practice in our institution, and analyzed incidents from an anonymous database to better characterize the practice and substantiate these assertions. Methods: To find examples of problems with piggyback, or secondary infusions, we searched the Food and Drug Administration's on-line incident database for incidents involving piggybacks. As part of a task analysis, 19 senior nurses each programmed 2 of 4 different pumps for a simulated piggyback infusion. To characterize infusion practice, we evaluated data logs from 55 infusion devices used in our institution. Results: Incidents from the database provided strong evidence that potential problems existed with piggyback infusions. Nurse behaviors suggested mismatches between the task, user, and devices that can lead to adverse events. Log files showed piggybacks were a common practice, and that available safeguards were not used. Conclusions: Our multiple data sources suggest automated piggyback infusion practice is neither simple nor safe. Incident report analysis suggests these findings contribute to adverse events. Further study is needed to understand and improve the safety of this practice.

This study sought to determine whether infusion device event logs could support accident investigation. Methods: An incident reporting database was searched for information about log file use in investigations. Log file data from devices in clinical use were downloaded and electronically searched for characteristics (signatures) matching specific function queries. Different programming sequences were simulated, and device logs were downloaded for analysis. Results: Database reports mentioned difficulties resolving log file data to the incident report and used log file data to confirm programming failures. Log file search revealed that, aside from alarm types and times, the devices were unable to adequately satisfy functional queries. Different simulated programming scenarios could not be easily differentiated by log file analysis. Conclusions: The device logs we studied collect data that are poorly suited to accident investigation. We conclude that infusion device logs cannot function as black boxes do in aviation accidents. Logs would be better applied to assist routine operations.

User queries to search engines are observed to predominantly contain inflected content words but lack stopwords and capitalization. Thus, they often resemble natural language queries after case folding and stopword removal. Query recovery aims to generate a linguistically well-formed query from a given user query as input to provide natural language processing tasks and cross-language information retrieval (CLIR). The evaluation of query translation shows that translation scores (NIST and BLEU) decrease after case folding, stopword removal, and stemming. A baseline method for query recovery reconstructs capitalization and stopwords, which considerably increases translation scores and significantly increases mean average precision for a standard CLIR task.

Learning environments aim to deliver efficacious instruction, but rarely take into consideration the motivational factors involved in the learning process. However, motivational aspects like engagement play an important role in effective learning-engaged learners gain more. E-Learning systems could be improved by tracking students' disengagement that, in turn, would allow personalized interventions at appropriate times in order to reengage students. This idea has been exploited several times for Intelligent Tutoring Systems, but not yet in other types of learning environments that are less structured. To address this gap, our research looks at online learning-content-delivery systems using educational data mining techniques. Previously, several attributes relevant for disengagement prediction were identified by means of log-file analysis on HTML-Tutor, a web-based learning environment. In this paper, we investigate the extendibility of our approach to other systems by studying the relevance of these attributes for predicting disengagement in a different e-learning system. To this end, two validation studies were conducted indicating that the previously identified attributes are pertinent for disengagement prediction, and two new meta-attributes derived from log-data observations improve prediction and may potentially be used for automatic log-file annotation.

Motivation-To measure computer users' personality from their use of keyboard and mouse. Research approach-Two explorative studies were conducted. In first study a background application was executed on 20 participants' computers to record keys pressed and mouse clicks on an average of eight days. In a second Study 15 participants' completed a programming task in an hour while a background application recorded keys pressed and mouse clicks. Participants were asked to complete the short form of IPIP-NEO personality inventory afterwards. Pearson correlation analysis was done between participants' behaviour on keyboard, mouse events and personality ratings. Findings/Design-The results suggest that some of the main traits and sub traits of personality can be measured from keyboard and mouse use. Significant correlations were found between personality main traits and sub traits and the use of keyboard and mouse. The maximum and minimum significant correlations were r (20) = 0.62, p <= 0.01 and r (20) = 0.40, p <= 0.05 respectively in first study and were r (15) = 0.7, p < 0.01 and r (15) = 0.51, p <= 0.05 respectively in the second study. Personality trait activity level was found to be significantly correlated with behavioural measure 'standard deviation of average time between events' in both the studies with r (20) = 0.54, p <= 0.05 and r (15) = 0.58, p <= 0.05 respectively Research limitations/Implications-The group of participants in both studies was relatively small. There was about 40% overlap of participants in the studies. Originality/Value-Relatively little research has focussed on personality and behaviour and specifically on the measurement of personality from the user behaviour on keyboard and mouse. Take away message-It might be possible to measure computer users' personality from their use of keyboard and mouse.

The aim of the research was to analyze the possibility of developing an Integrated Development Environment (IDE) that could improve a programmer's performance by considering their current mood. Various experiments were conducted to study this idea. However, the impact of moods on programmer performance was initially examined in the literature. Based on this, a Cognitive Programming Task Model (CPTM) was developed showing that various cognitive functions and programming activities are interrelated. A second model derived from the literature, the Cognitive Mood Model (CMM), suggested that moods are also interrelated with various cognitive functions. Combining these two models indirectly suggests a relation between moods and programming tasks, which was presented as the Mood Programming Model (MPM). As direct empirical support was lacking for this relation, two experiments were conducted to study the effect mood could have on performance in a debug task. Validated mood-inducing movie clips were used to induce specific moods along twomood dimensions: valence and arousal. The first study was conducted online. The results showed that arousal is a significant factor when considering programmer performance whereas valence was found to have no significant effect. The second study was a continuation study to validate the findings from the first study within lab conditions. The results were not able to confirm the findings of the first experiment. The reasons for these findings were explained accordingly. As mood was found to have an effect on a programmer's coding and debugging performance, this factor might be considered when developing a support system. The next step in the research was therefore to consider mood measuring in a non-interruptive way. The next two experiments were based around the hypothesis that-moods can be measured from the keyboard and mouse interaction of the computer user‖. In the first experiment an application was installed on participants' computers to record their key presses and mouse clicks in a log file. Their self reported moods in intervals of 20 minutes were also stored in the same file over an average period of eight days. Correlations between participants' self reported moods and their keyboard and mouse use revealed that it might be possible to measure moods of the some of the participants. The second experiment took place in the lab, where participants were asked to perform programming like tasks while listening to Declaration The following publications have been produced, directly and indirectly, from the research undertaken in this investigation.

The purpose of this explorative research was to explore the mood of a computer user and his or her use of keyboard and mouse. Twenty-six users (13 programmers and 13 frequent computer users) took part in the study. A background software application executing on participants&#x27; computers logged the keyboard key press and mouse click events. The correlations between moods of the participants and their use of keyboard and mouse show that it might be possible to create individual tailor made mood measures ...

In an intelligent tutoring system (ITS), it can be useful to know when a student has disengaged from a task and might benefit from a particular intervention. However, predicting disengagement on a trial-by-trial basis is a challenging problem, particularly in complex cognitive domains. In the present work, data-driven methods were used to address two aspects of this problem: identification of predictive features at the single-trial level, and selection of accurate and robust models. Experiment data were collected in a middle-school classroom using a vocabulary training ITS. On each trial, the ITS presented a low-frequency (Tier 2 or frontier) word and prompted students to type in the word's meaning. Single-trial measures-including the orthographic and semantic accuracy of each response, and context-sensitive measures such as interaction patterns across trials-were computed throughout the task. There were two key findings. First, as expected, different features predicted when a student was likely to be more engaged (e.g., high semantic accuracy) or less engaged (e.g., repetition of same or similar words across consecutive trials). Second, there was added value in representing context-sensitive information, which captures patterns of performance over time, as well as trial-specific information. These findings provide useful insights into effective methods for representing and modeling temporal patterns of student engagement in an ITS, especially those related to language learning. Such models may be useful in the design and implementation of adaptive tutors in complex cognitive domains like language learning.

While much has been written about the importance of innovation, there is still much to learn about the specific behaviours that lead to innovation among groups. In this paper we introduce a framework of innovation based on behaviours identified as being conducive to collaborative innovation. We also report on a study of a task designed to elicit innovation supportive behaviours in a virtual world environment. The task resulted in a variety of solutions and a range of participant behaviours, and specific behaviours were correlated with innovative solutions. Multiple forms of analysis provided unique insights into participant behaviour, and the combined set of analyses led to a richer understanding of participant behaviour than found through any individual analysis. The paper also presents implications for how organizations may scaffold group interactions to increase the chances of successful collaborative innovation.

Multi-touch tabletops are seen as devices that encourage equity in collaborative work: everyone can join in, either by talking or doing, and people who say less sometimes tend to do more (Harris et al, 2009; Rogers et al., 2009). However, is it enough to consider equal contributions (verbal or physical) as measures of equity? In our work investigating how technology might encourage collaboration, we have been forced to question what ‘good ’ collaboration is. Adults using tabletops certainly tend to perceive them as encouraging greater equity compared to mouse input, for example (Marshall et al., 2009). But when making a good collaborative decision, is it really possible to say whose decision it was? In young children, close collaboration is marked by the ‘I did it! ’ bias: the tendency to attribute to oneself the actions of others, through ‘appropriating the mental experience ’ of the partner (Rogoff, 1990). In educational settings, a good tutor may often try to make an idea the chi...

This paper presents a classroom study that investigated the potential of using touch tabletop technology to support children’s collaborative learning interactions. Children aged 710 worked in groups of three on a collaborative planning task in which they designed a seating plan for their classroom. In the single-touch condition, the tabletop surface allowed only one child to interact with the digital content at a time. In the multiple-touch condition, the children could interact with the digital content simultaneously. Results showed that touch condition did not affect the frequency or equity of interactions, but did influence the nature of children’s discussion. In the multiple-touch condition, children talked more about the task; in the singletouch condition, they talked more about turn taking. We also report age and gender differences.