A New Android Malware Detection Method Using Bayesian Classification

Mobile malware has been growing in scale and complexity as smartphone usage continues to rise. Android has surpassed other mobile platforms as the most popular whilst also witnessing a dramatic increase in malware targeting the platform. A worrying trend that is emerging is the increasing sophistication of Android malware to evade detection by traditional signature-based scanners. As such, Android app marketplaces remain at risk of hosting malicious apps that could evade detection before being downloaded by unsuspecting users. Hence, in this paper we present an effective approach to alleviate this problem based on Bayesian classification models obtained from static code analysis. The models are built from a collection of code and app characteristics that provide indicators of potential malicious activities. The models are evaluated with real malware samples in the wild and results of experiments are presented to demonstrate the effectiveness of the proposed approach.

The Impact of Colleague Peer Review on the Radiotherapy Treatment Planning Process in the Radical Treatment of Lung Cancer

AIMS: Modern radiotherapy uses techniques to reliably identify tumour and reduce target volume margins. However, this can potentially lead to an increased risk of geographic miss. One source of error is the accuracy of target volume delineation (TVD). Colleague peer review (CPR) of all curative-intent lung cancer plans has been mandatory in our institution since May 2013. At least two clinical oncologists review plans, checking treatment paradigm, TVD, prescription dose tumour and critical organ tolerances. We report the impact of CPR in our institution.

MATERIALS AND METHODS: Radiotherapy treatment plans of all patients receiving radical radiotherapy were presented at weekly CPR meetings after their target volumes were reviewed and signed off by the treating consultant. All cases and any resultant change to TVD (including organs at risk) or treatment intent were recorded in our prospective CPR database. The impact of CPR over a 13 month period from May 2013 to June 2014 is reported.

RESULTS: One hundred and twenty-two patients (63% non-small cell lung carcinoma, 17% small cell lung carcinoma and 20% 'clinical diagnosis') were analysed. On average, 3.2 cases were discussed per meeting (range 1-8). CPR resulted in a change in treatment paradigm in 3% (one patient proceeded to induction chemotherapy, two patients had high-dose palliative radiotherapy). Twenty-one (17%) had a change in TVD and one (1%) patient had a change in dose prescription. In total, 6% of patients had plan adjustment after review of dose volume histogram.

CONCLUSION: The introduction of CPR in our centre has resulted in a change in a component of the treatment plan for 27% of patients receiving curative-intent lung radiotherapy. We recommend CPR as a mandatory quality assurance step in the planning process of all radical lung plans.

A field and reactive transport model study of arsenic in a basaltic rock aquifer

The use of geothermal energy as a source for electricity and district heating has increased over recent decades. Dissolved As can be an important constituent of the geothermal fluids brought to the Earth's surface. Here the field application of laboratory measured adsorption coefficients of aqueous As species on basaltic glass surfaces is discussed. The mobility of As species in the basaltic aquifer in the Nesjavellir geothermal system, Iceland was modelled by the one-dimensional (1D) reactive transport model PHREEQC ver. 2, constrained by a long time series of field measurements with the chemical composition of geothermal effluent fluids, pH, Eh and, occasionally, Fe- and As-dissolved species measurements. Di-, tri- and tetrathioarsenic species (As(OH)S22-, AsS₃H^2-, AsS33- and As(SH)4-) were the dominant form of dissolved As in geothermal waters exiting the power plant (2.556μM total As) but converted to some extent to arsenite (H₃AsO₃) and arsenate HAsO42- oxyanions coinciding with rapid oxidation of S2- to S2O32- and finally to SO42- during surface runoff before feeding into a basaltic lava field with a total As concentration of 0.882μM following dilution with other surface waters. A continuous 25-a data set monitoring groundwater chemistry along a cross section of warm springs on the Lake Thingvallavatn shoreline allowed calibration of the 1D model. Furthermore, a series of ground water wells located in the basaltic lava field, provided access along the line of flow of the geothermal effluent waters towards the lake. The conservative ion Cl^- moved through the basaltic lava field (4100m) in less than10a but As was retarded considerably due to surface reactions and has entered a groundwater well 850m down the flow path as arsenate in accordance to the prediction of the 1D model. The 1D model predicted a complete breakthrough of arsenate in the year 2100. In a reduced system arsenite should be retained for about 1ka. © 2011 Elsevier Ltd.

High accuracy android malware detection using ensemble learning

With over 50 billion downloads and more than 1.3 million apps in Google’s official market, Android has continued to gain popularity amongst smartphone users worldwide. At the same time there has been a rise in malware targeting the platform, with more recent strains employing highly sophisticated detection avoidance techniques. As traditional signature based methods become less potent in detecting unknown malware, alternatives are needed for timely zero-day discovery. Thus this paper proposes an approach that utilizes ensemble learning for Android malware detection. It combines advantages of static analysis with the efficiency and performance of ensemble machine learning to improve Android malware detection accuracy. The machine learning models are built using a large repository of malware samples and benign apps from a leading antivirus vendor. Experimental results and analysis presented shows that the proposed method which uses a large feature space to leverage the power of ensemble learning is capable of 97.3 % to 99% detection accuracy with very low false positive rates.