Seletor de Idiomas

Top - Webmail / Login / Trac

Imagem - Destaques

slide interno 8

Technical Session 2 - AI and Intelligent Services

Title: A Counselors-Based Intrusion Detection Architecture

Author: Silvio Ereno Quincozes


Abstract: Intrusion Detection Systems (IDSs) are a fundamental component of defensive solutions. In particular, signature-based IDSs aim to detect malicious activities on computer systems and networks by relying on data classification models built from a training dataset. However, classifiers performance can vary for each attack pattern. A common technique to overcome this issue is to use ensemble methods, where multiple classifiers are employed and a final decision is taken combining their outputs. Despite the potential advantages of such an approach, its usefulness is limited in scenarios where (i) multiple expert classifiers present divergent results or (ii) representative data are missing to detect a specific attack class. In this work, we introduce the concept of counselor networks to deal with conflicts from different classifiers by exploiting the collaboration between IDSs that analyze multiple and heterogeneous data sources. Our empirical results demonstrate the feasibility of the proposed architecture in improving the accuracy of the intrusion detection process.




Title: A Cache Prefetch Policy based on Users’ Temporal-and-Social Behavior for Content Management in Wireless Access Networks

Author: Cleomar Márcio Marques de Oliveira


Abstract: In this article, a new policy is proposed for store and drops cache content in the Wireless Access Networks nodes. The proposed policy select content that can be dropped and new content to be cached in a network node, on predefined time periods at each day and with pre-established time duration each one, repeated at each day. The temporal aspects and users social behavior that connect to the node for decision making are considered. An algorithm selects new content, in the same proportion, from those categories historically requested ones in these time periods on previous day. These new content selection purpose is to cache them in current day, at the corresponding time periods, to increase the content request hit ratio according to this policy. Simulation results against established FIFO, LRU, LFU and RANDOM policies show that this proposed policy hit ratio is 2.46 times higher than the others, with an average hit ratio of 13.1% here versus an average hit ratio 5.325% of the above cited policies, in the evaluated scenarios.




Title: Using ubus for collecting data and remote configuration of OpenWRT Access Points

Author: Yago de Rezende dos Santos


Abstract: As SCIFI progressed, different protocols were used for collecting data from and configuring Access Points. Originally ssh and scp were used for increased security on an open network. When the control network was isolated on it's own vlan, data collection and configuration migrated to snmp. Finally, the new design calls for rpc using the ubus infrastructure.





Title: Natural Language Processing Characterization of Recurring Calls in Public Security Services

Author: Nicollas Rodrigues De Oliveira


Abstract: Extracting knowledge from unstructured data silos, a legacy of old applications, is mandatory for improving the governance of today's cities and fostering the creation of smart cities. Texts in natural language often compose such data. Nevertheless, the inference of useful information from a linguistic-computational analysis of natural language data is an open challenge. In this paper, we propose a clustering method to analyze textual data employing the unsupervised machine learning algorithms k-means and hierarchical clustering. We assess different vector representation methods for text, similarity metrics, and the number of clusters that best matches the data. We evaluate the methods using a real database of a public record service of security occurrences. The results show that the k-means algorithm using Euclidean distance extracts non-trivial knowledge, reaching up to 93% accuracy in a set of test samples while identifying the 12 most prevalent occurrence patterns.

Social - Facebook - Twitter - G+