5 resultados para Cluster Analysis. Information Theory. Entropy. Cross Information Potential. Complex Data
em Universitätsbibliothek Kassel, Universität Kassel, Germany
Resumo:
Die zunehmende Vernetzung der Informations- und Kommunikationssysteme führt zu einer weiteren Erhöhung der Komplexität und damit auch zu einer weiteren Zunahme von Sicherheitslücken. Klassische Schutzmechanismen wie Firewall-Systeme und Anti-Malware-Lösungen bieten schon lange keinen Schutz mehr vor Eindringversuchen in IT-Infrastrukturen. Als ein sehr wirkungsvolles Instrument zum Schutz gegenüber Cyber-Attacken haben sich hierbei die Intrusion Detection Systeme (IDS) etabliert. Solche Systeme sammeln und analysieren Informationen von Netzwerkkomponenten und Rechnern, um ungewöhnliches Verhalten und Sicherheitsverletzungen automatisiert festzustellen. Während signatur-basierte Ansätze nur bereits bekannte Angriffsmuster detektieren können, sind anomalie-basierte IDS auch in der Lage, neue bisher unbekannte Angriffe (Zero-Day-Attacks) frühzeitig zu erkennen. Das Kernproblem von Intrusion Detection Systeme besteht jedoch in der optimalen Verarbeitung der gewaltigen Netzdaten und der Entwicklung eines in Echtzeit arbeitenden adaptiven Erkennungsmodells. Um diese Herausforderungen lösen zu können, stellt diese Dissertation ein Framework bereit, das aus zwei Hauptteilen besteht. Der erste Teil, OptiFilter genannt, verwendet ein dynamisches "Queuing Concept", um die zahlreich anfallenden Netzdaten weiter zu verarbeiten, baut fortlaufend Netzverbindungen auf, und exportiert strukturierte Input-Daten für das IDS. Den zweiten Teil stellt ein adaptiver Klassifikator dar, der ein Klassifikator-Modell basierend auf "Enhanced Growing Hierarchical Self Organizing Map" (EGHSOM), ein Modell für Netzwerk Normalzustand (NNB) und ein "Update Model" umfasst. In dem OptiFilter werden Tcpdump und SNMP traps benutzt, um die Netzwerkpakete und Hostereignisse fortlaufend zu aggregieren. Diese aggregierten Netzwerkpackete und Hostereignisse werden weiter analysiert und in Verbindungsvektoren umgewandelt. Zur Verbesserung der Erkennungsrate des adaptiven Klassifikators wird das künstliche neuronale Netz GHSOM intensiv untersucht und wesentlich weiterentwickelt. In dieser Dissertation werden unterschiedliche Ansätze vorgeschlagen und diskutiert. So wird eine classification-confidence margin threshold definiert, um die unbekannten bösartigen Verbindungen aufzudecken, die Stabilität der Wachstumstopologie durch neuartige Ansätze für die Initialisierung der Gewichtvektoren und durch die Stärkung der Winner Neuronen erhöht, und ein selbst-adaptives Verfahren eingeführt, um das Modell ständig aktualisieren zu können. Darüber hinaus besteht die Hauptaufgabe des NNB-Modells in der weiteren Untersuchung der erkannten unbekannten Verbindungen von der EGHSOM und der Überprüfung, ob sie normal sind. Jedoch, ändern sich die Netzverkehrsdaten wegen des Concept drif Phänomens ständig, was in Echtzeit zur Erzeugung nicht stationärer Netzdaten führt. Dieses Phänomen wird von dem Update-Modell besser kontrolliert. Das EGHSOM-Modell kann die neuen Anomalien effektiv erkennen und das NNB-Model passt die Änderungen in Netzdaten optimal an. Bei den experimentellen Untersuchungen hat das Framework erfolgversprechende Ergebnisse gezeigt. Im ersten Experiment wurde das Framework in Offline-Betriebsmodus evaluiert. Der OptiFilter wurde mit offline-, synthetischen- und realistischen Daten ausgewertet. Der adaptive Klassifikator wurde mit dem 10-Fold Cross Validation Verfahren evaluiert, um dessen Genauigkeit abzuschätzen. Im zweiten Experiment wurde das Framework auf einer 1 bis 10 GB Netzwerkstrecke installiert und im Online-Betriebsmodus in Echtzeit ausgewertet. Der OptiFilter hat erfolgreich die gewaltige Menge von Netzdaten in die strukturierten Verbindungsvektoren umgewandelt und der adaptive Klassifikator hat sie präzise klassifiziert. Die Vergleichsstudie zwischen dem entwickelten Framework und anderen bekannten IDS-Ansätzen zeigt, dass der vorgeschlagene IDSFramework alle anderen Ansätze übertrifft. Dies lässt sich auf folgende Kernpunkte zurückführen: Bearbeitung der gesammelten Netzdaten, Erreichung der besten Performanz (wie die Gesamtgenauigkeit), Detektieren unbekannter Verbindungen und Entwicklung des in Echtzeit arbeitenden Erkennungsmodells von Eindringversuchen.
Resumo:
Formal Concept Analysis allows to derive conceptual hierarchies from data tables. Formal Concept Analysis is applied in various domains, e.g., data analysis, information retrieval, and knowledge discovery in databases. In order to deal with increasing sizes of the data tables (and to allow more complex data structures than just binary attributes), conceputal scales habe been developed. They are considered as metadata which structure the data conceptually. But in large applications, the number of conceptual scales increases as well. Techniques are needed which support the navigation of the user also on this meta-level of conceptual scales. In this paper, we attack this problem by extending the set of scales by hierarchically ordered higher level scales and by introducing a visualization technique called nested scaling. We extend the two-level architecture of Formal Concept Analysis (the data table plus one level of conceptual scales) to many-level architecture with a cascading system of conceptual scales. The approach also allows to use representation techniques of Formal Concept Analysis for the visualization of thesauri and ontologies.
Resumo:
Social resource sharing systems like YouTube and del.icio.us have acquired a large number of users within the last few years. They provide rich resources for data analysis, information retrieval, and knowledge discovery applications. A first step towards this end is to gain better insights into content and structure of these systems. In this paper, we will analyse the main network characteristics of two of the systems. We consider their underlying data structures – socalled folksonomies – as tri-partite hypergraphs, and adapt classical network measures like characteristic path length and clustering coefficient to them. Subsequently, we introduce a network of tag co-occurrence and investigate some of its statistical properties, focusing on correlations in node connectivity and pointing out features that reflect emergent semantics within the folksonomy. We show that simple statistical indicators unambiguously spot non-social behavior such as spam.
Resumo:
Social resource sharing systems like YouTube and del.icio.us have acquired a large number of users within the last few years. They provide rich resources for data analysis, information retrieval, and knowledge discovery applications. A first step towards this end is to gain better insights into content and structure of these systems. In this paper, we will analyse the main network characteristics of two of these systems. We consider their underlying data structures â so-called folksonomies â as tri-partite hypergraphs, and adapt classical network measures like characteristic path length and clustering coefficient to them. Subsequently, we introduce a network of tag cooccurrence and investigate some of its statistical properties, focusing on correlations in node connectivity and pointing out features that reflect emergent semantics within the folksonomy. We show that simple statistical indicators unambiguously spot non-social behavior such as spam.
Resumo:
Parasitic weeds of the genera Striga, Orobanche, and Phelipanche pose a severe problem for agriculture because they are difficult to control and are highly destructive to several crops. The present work was carried out during the period October, 2009 to February, 2012 to evaluate the potential of arbuscular mycorrhizal fungi (AMF) to suppress P. ramosa on tomatoes and to investigate the effects of air-dried powder and aqueous extracts from Euphorbia hirta on germination and haustorium initiation in Phelipanche ramosa. The work was divided into three parts: a survey of the indigenous mycorrhizal flora in Sudan, second, laboratory and greenhouse experiments (conducted in Germany and Sudan) to construct a base for the third part, which was a field trial in Sudan. A survey was performed in 2009 in the White Nile state, Sudan to assess AMF spore densities and root colonization in nine fields planted with 13 different important agricultural crops. In addition, an attempt was made to study the relationship between soil physico-chemical properties and AMF spore density, colonization rate, species richness and other diversity indices. The mean percentage of AMF colonization was 34%, ranging from 19-50%. The spore densities (expressed as per 100 g dry soil) retrieved from the rhizosphere of different crops were relatively high, varying from 344 to 1222 with a mean of 798. There was no correlation between spore densities in soil and root colonization percentage. A total of 45 morphologically classifiable species representing ten genera of AMF were detected with no correlation between the number of species found in a soil sample and the spore density. The most abundant genus was Glomus (20 species). The AMF diversity expressed by the Shannon–Weaver index was highest in sorghum (H\= 2.27) and Jews mallow (H\= 2.13) and lowest in alfalfa (H\= 1.4). With respect to crop species, the genera Glomus and Entrophospora were encountered in almost all crops, except for Entrophospora in alfalfa. Kuklospora was found only in sugarcane and sorghum. The genus Ambispora was recovered only in mint and okra, while mint and onion were the only species on which no Acaulospora was found. The hierarchical cluster analysis based on the similarity among AMF communities with respect to crop species overall showed that species compositions were relatively similar with the highest dissimilarity of about 25% separating three of the mango samples and the four sorghum samples from all other samples. Laboratory experiments studied the influence of root and stem exudates of three tomato varieties infected by three different Glomus species on germination of P. ramosa. Root exudates were collected 21or 42 days after transplanting (DAT) and stem exudates 42 DAT and tested for their effects on germination of P. ramosa seeds in vitro. The tomato varieties studied did not have an effect on either mycorrhizal colonization or Phelipanche germination. Germination in response to exudates from 42 day old mycorrhizal plants was significantly reduced in comparison to non-mycorrhizal controls. Germination of P. ramosa in response to root exudates from 21 day old plants was consistently higher than for 42 day-old plants (F=121.6; P<.0001). Stem diffusates from non-mycorrhizal plants invariably elicited higher germination than diffusates from the corresponding mycorrhizal ones and differences were mostly statistically significant. A series of laboratory experiments was undertaken to investigate the effects of aqueous extracts from Euphorbia hirta on germination, radicle elongation, and haustorium initiation in P. ramosa. P. ramosa seeds conditioned in water and subsequently treated with diluted E. hirta extract (10-25% v/v) displayed considerable germination (47-62%). Increasing extract concentration to 50% or more reduced germination in response to the synthetic germination stimulants GR24 and Nijmegen-1 in a concentration dependent manner. P. ramosa germlings treated with diluted Euphorbia extract (10-75 % v/v) displayed haustorium initiation comparable to 2, 5-Dimethoxy-p-benzoquinon (DMBQ) at 20 µM. Euphorbia extract applied during conditioning reduced haustorium initiation in a concentration dependent manner. E. hirta extract or air-dried powder, applied to soil, induced considerable P. ramosa germination. Pot experiments were undertaken in a glasshouse at the University of Kassel, Germany, to investigate the effects of P. ramosa seed bank on tomato growth parameters. Different Phelipanche seed banks were established by mixing the parasite seeds (0 - 32 mg) with the potting medium in each pot. P. ramosa reduced all tomato growth parameters measured and the reduction progressively increased with seed bank. Root and total dry matter accumulation per tomato plant were most affected. P. ramosa emergence, number of tubercles, and tubercle dry weight increased with the seed bank and were, invariably, maximal with the highest seed bank. Another objective was to determine if different AM fungi differ in their effects on the colonization of tomatoes with P. ramosa and the performance of P. ramosa after colonization. Three AMF species viz. GIomus intraradices, Glomus mosseae and Glomus Sprint® were used in this study. For the infection, P. ramosa seeds (8 mg) were mixed with the top 5 cm soil in each pot. No mycorrhizal colonization was detected in un-inoculated control plants. P. ramosa infested, mycorrhiza inoculated tomato plants had significantly lower AMF colonization compared to plants not infested with P. ramosa. Inoculation with G. intraradices, G. mosseae and Glomus Sprint® reduced the number of emerged P. ramosa plants by 29.3, 45.3 and 62.7% and the number of tubercles by 22.2, 42 and 56.8%, respectively. Mycorrhizal root colonization was positively correlated with number of branches and total dry matter of tomatoes. Field experiments on tomato undertaken in 2010/12 were only partially successful because of insect infestations which resulted in the complete destruction of the second run of the experiment. The effects of the inoculation with AMF, the addition of 10 t ha-1 filter mud (FM), an organic residues from sugar processing and 36 or 72 kg N ha-1 on the infestation of tomatoes with P. ramosa were assessed. In un-inoculated control plants, AMF colonization ranged between 13.4 to 22.1% with no significant differences among FM and N treatments. Adding AMF or FM resulted in a significant increase of branching in the tomato plants with no additive effects. Dry weights were slightly increased through FM application when no N was applied and significantly at 36 kg N ha-1. There was no effect of FM on the time until the first Phelipanche emerged while AMF and N application interacted. Especially AMF inoculation resulted in a tendency to delayed P. ramosa emergence. The marketable yield was extremely low due to the strong fruit infestation with insects mainly whitefly Bemisia tabaci and tomato leaf miner (Tuta absoluta). Tomatoes inoculated with varied mycorrhiza species displayed different response to the insect infestation, as G. intraradices significantly reduced the infestation, while G. mosseae elicited higher insect infestation. The results of the present thesis indicate that there may be a potential of developing management strategies for P. ramosa targeting the pre-attachment stage namely germination and haustorial initiation using plant extracts. However, ways of practical use need to be developed. If such treatments can be combined with AMF inoculation also needs to be investigated. Overall, it will require a systematic approach to develop management tools that are easily applicable and affordable to Sudanese farmers. It is well-known that proper agronomical practices such as the design of an optimum crop rotation in cropping systems, reduced tillage, promotion of cover crops, the introduction of multi-microbial inoculants, and maintenance of proper phosphorus levels are advantageous if the mycorrhiza protection method is exploited against Phelipanche ramosa infestation. Without the knowledge about the biology of the parasitic weeds by the farmers and basic preventive measures such as hygiene and seed quality control no control strategy will be successful, however.