Christoph Schommer

Managing conversational streams by explorative mind-maps

In this paper, we introduce an explorative but adaptive-associative information management system in the presence of a natural conversation. We take advantage of explorative mind-maps, which have been demonstrated in [10] and which are altogether a management framework that emerges automatically from the data input stream it gets. An explorative mind-map is a non-verificative but dynamic system that basis on the natural paradigm: it changes its complexity continuously and fosters symbolic cells according to internal activation states. Generally, the structure mirrors a mental state where the oblivion of associated facts arrive once the stimulation decreases. Considering two mind-maps A1,2 and B1,2 for two conversational partners A and B, the mind-map *1 represents the self-conversation and *2 the conversational stream of the conversational partner. If we merge these mind-maps, we may apply the out-coming results for the computation of trust.

A Cognitive Mind-Map Framework to Foster Trust

The explorative mind-map is a dynamic framework, that emerges automatically from the input, it gets. It is unlike a verificative modeling system where existing (human) thoughts are placed and connected together. In this regard, explorative mind-maps change their size continuously, being adaptive with connectionist cells inside; mind-maps process data input incrementally and offer lots of possibilities to interact with the user through an appropriate communication interface. With respect to a cognitive motivated situation like a conversation between partners, mind-maps become interesting as they are able to process stimulating signals whenever they occur. If these signals are close to an own understanding of the world, then the conversational partner becomes automatically more trustful than if the signals do not or less match the own knowledge scheme. In this (position) paper, we therefore motivate explorative mind-maps as a cognitive engine and propose these as a decision support engine to foster trust.

e-Conviviality in web systems by the wisdom of crowds

We motivate the idea of e-conviviality in web-based systems and argue that a convivial social being deeply depends on the implicit and explicit co-operation and collaboration of natural users inside a community. We believe that a (individual) conviviality benefits from the wisdom of crowds, meaning that a continuously and dynamic understanding of the users behaviour can heavily influence the individual well being. In this respect, we are currently implementing the system CUBA (conviviality and user behaviour analysis), which aims to find novel ways to support users during their web site visits while discovering their interests. CUBA comes up with certain recommendations and suggestions, which base on a common behaviour of the “Wisdom of Crowds”. For example, concepts with respect to time, space, and user-based actions are considered.

Integration of Manual and Automatic Text Categorization. A Categorization Workbench for Text-Based Email and Spam

As a method structuring information and knowledge contained in texts, text categorization can be to a great extend automated. The automatic text classification systems implement machine learning algorithms and need training samples. In commercial applications however, the automatic categorization appear to come up against limiting factors. For example, it turns out to be difficult to reduce the sample complexity without the categorization quality in terms of recall and precision will suffer. Instead of trying to fully replace the human work by machine, it could be more effective and ultimately efficient to let human and machine cooperate. So we have developed a categorization workbench to realise synergy between manual and machine categorization. To compare the categorization workbench with common automatic classification systems, the automatic categorizer of the IBM DB2 Information Integrator for Content has been chosen for tests. The test results show that, benefiting from the incorporation of user’s domain knowledge, the categorization workbench can improve the recall by a factor of two till four with the same number of training samples as the automatic categorizer uses. Further, to get a comparable categorization quality, the categorization workbench just needs an eighth till a quarter of the training samples as the automatic categorizer does.

PERSEUS: A Personalization Framework for Sentiment Categorization with Recurrent Neural Network

This paper introduces the personalization framework PERSEUS in order to investigate the impact of individuality in sentiment categorization by looking into the past. The existence of diversity between individuals and certain consistency in each individual is the cornerstone of the framework. We focus on relations between documents for user-sensitive predictions. Individual’s lexical choices act as indicators for individuality, thus we use a concept-based system which utilizes neural networks to embed concepts and associated topics in text. Furthermore, a recurrent neural network is used to memorize the history of user’s opinions, to discover user-topic dependence, and to detect implicit relations between users. PERSEUS also offers a solution for data sparsity. At the first stage, we show the benefit of inquiring a user-specified system. Improvements in performance experimented on a combined Twitter dataset are shown over generalized models. PERSEUS can be used in addition to such generalized systems to enhance the understanding of user’s opinions.

Psychological, cognitive factors and contextual influences in pain and pain-related suffering as revealed by a combined qualitative and quantitative assessment approach

Previous psychophysiological research suggests that pain measurement needs to go beyond the assessment of Pain Intensity and Unpleasantness by adding the evaluation of Pain-Related Suffering. Based on this three-dimensional approach, we attempted to elucidate who is more likely to suffer by identifying reasons that may lead individuals to report Pain and Pain-Related Suffering more than others. A sample of 24 healthy participants (age range 18–33) underwent four different sessions involving the evaluation of experimentally induced phasic and tonic pain. We applied two decision tree models to identify variables (selected from psychological questionnaires regarding pain and descriptors from post-session interviews) that provided a qualitative characterization of the degrees of Pain Intensity, Unpleasantness and Suffering and assessed the respective impact of contextual influences. The overall classification accuracy of the decision trees was 75% for Intensity, 77% for Unpleasantness and 78% for Pain-Related Suffering. The reporting of suffering was predominantly associated with fear of pain and active cognitive coping strategies, pain intensity with bodily competence conveying strength and resistance and unpleasantness with the degree of fear of pain and catastrophizing. These results indicate that the appraisal of the three pain dimensions was largely determined by stable psychological constructs. They also suggest that individuals manifesting higher active coping strategies may suffer less despite enhanced pain and those who fear pain may suffer even under low pain. The second decision tree model revealed that suffering did not depend on pain alone, but that the complex rating-related decision making can be shifted by situational factors (context, emotional and cognitive). The impact of coping and fear of pain on individual Pain-Related Suffering may highlight the importance of improving cognitive coping strategies in clinical settings.

Towards the Identification of Outliers in Satellite Telemetry Data by Using Fourier Coefficients

Spacecrafts provide a large set of on-board components information such as their temperature, power and pressure. This information is constantly monitored by engineers, who capture the outliers and determine whether the situation is abnormal or not. However, due to the large quantity of information, only a small part of the data is being processed or used to perform anomaly early detection. A common accepted research concept for anomaly prediction as described in literature yields on using projections, based on probabilities, estimated on learned patterns from the past [6] and data mining methods to enhance the conventional diagnosis approach [14]. Most of them conclude on the need to build a pattern identity chart. We propose an algorithm for efficient outlier detection that builds an identity chart of the patterns using the past data based on their curve fitting information. It detects the functional units of the patterns without apriori knowledge with the intent to learn its structure and to reconstruct the sequence of events described by the signal. On top of statistical elements, each pattern is allotted a characteristics chart. This pattern identity enables fast pattern matching across the data. The extracted features allow classification with regular clustering methods like support vector machines SVM. The algorithm has been tested and evaluated using real satellite telemetry data. The outcome and performance show promising results for faster anomaly prediction.

Finding Outliers in Satellite Patterns by Learning Pattern Identities

Spacecrafts provide a large set of on-board components information such as their temperature, power and pressure. This information is constantly monitored by engineers, who capture the outliers and determine whether the situation is abnormal or not. However, due to the large quantity of information, only a small part of the data is being processed or used to perform anomaly prediction. A common accepted research concept for anomaly prediction as described in literature yields on using projections, based on probabilities, estimated on learned patterns from the past (Fujimaki et al., 2005) and data mining methods to enhance the conventional diagnosis approach (Li et al., 2010). Most of them conclude on the need to build a status vector. We propose an algorithm for efficient outlier detection that builds an identity chart of the patterns using the past data based on their curve fitting information. It detects the functional units of the patterns without apriori knowledge with the intent to learn its structure and to reconstruct the sequence of events described by the signal. On top of statistical elements, each pattern is allotted a characteristics chart. This pattern identity enables fast pattern matching across the data. The extracted features allow classification with regular clustering methods like support vector machines (SVM). The algorithm has been tested and evaluated using real satellite telemetry data. The outcome and performance show promising results for faster anomaly prediction.

Domain-driven news representation using conditional attribute-value pairs

Financial news carry information about economical figures and indicators. However, these texts are mostly unstructured and consequently hard to be processed in an automatic way. In this paper, we present a representation formalism that supports a linguistic composition for machine learning tasks. We show an innovative approach to structuring financial texts by extracting principal indicators. Considering announcements in the monetary policy domain, we distinguish between attributes and their values and argue that attributes are to be represented as an aggregated set of economic terms, keeping their values as corresponding conditional expressions. We close with a critical discussion and future perspectives.

Sieving publishing communities in DBLP

DBLP is a bibliographic database with more than one million data entries, collected from the last 70 years, and labeled with diverse attributes like the authorspsila names, the publication title, and the year of publishing. With this as ground, the motivation of applying analytical examinations to identifying publishing communities become meaningful. In this respect and focusing on the idea of figuring out existing associative connectivity between authors, this work exposes some novel information as for example the most frequent community patterns, the ldquoDonald E. Ullmanrdquo-star of 1975, and an example for a typical extreme-sized community. We close with a temporal flight throughout the decades while observing the extreme-sized community and highlight perspectives and further analytical suggestions.