Social networks crawling is in the focus of active research the last years. One of the challenging task is to collect target nodes in an initially unknown graph given a budget of crawling steps. Predicting a node prop...
详细信息
Social networks crawling is in the focus of active research the last years. One of the challenging task is to collect target nodes in an initially unknown graph given a budget of crawling steps. Predicting a node property based on its partially known neighbourhood is at the heart of a successful crawler. In this paper we adopt graph neural networks for this purpose and show they are competitive to traditional classifiers and are better for individual cases. Additionally we suggest a training sample boosting technique, which helps to diversify the training set at early stages of crawling and thus improves the predictor quality. The experimental study on three types of target set topology indicates GNN based approach has a potential in crawling task, especially in the case of distributed target nodes.
In recent years, as data and problem sizes have increased, distributed learning has become an essential tool for training highperformance models. However, the communication bottleneck, especially for high-dimensional ...
详细信息
The semantics of the core features of XML Schema in terms of XQuery 1.0 and XPath 2.0 data model algebraically defined is given. The database state is represented as a many sorted algebra whose sorts are sets of data ...
详细信息
The semantics of the core features of XML Schema in terms of XQuery 1.0 and XPath 2.0 data model algebraically defined is given. The database state is represented as a many sorted algebra whose sorts are sets of data type values and different kinds of nodes and whose operations are data type operations and node accessors. The values of some node accessors, such as "parent", "children" and "attributes", define a document tree with a definite order of nodes. The values of other node accessors help to make difference between kinds of nodes, learn the names, types and values associated with the corresponding document entities, etc., i.e., provide primitive facilities for a query language. As a result, a document can be easily mapped to its implementation in terms of nodes and accessors defined on them.
By obfuscation we mean any efficient semantic-preserving transformation of computer programs aimed at bringing a program into such a form, which impedes the understanding of its algorithm and data structures or preven...
详细信息
Program's source code usually contains many calls of library functions without available sources, therefore there is no way to analyze them like user code. To achieve accuracy required by industrial static analyze...
Program's source code usually contains many calls of library functions without available sources, therefore there is no way to analyze them like user code. To achieve accuracy required by industrial static analyzer, various library modeling methods are used. The paper presents one of the methods used by the SharpChecker static source code analyzer for modeling library functions using C# specifications in the symbolic execution engine. The approach allows to store and explore predicates of events, important side-effects and other information useful for error detectors with reasonable accuracy. It makes it possible to model side-effects of user-defined lambda-functions passed to external methods as parameters, which is especially important for frequently used C# LINQ methods. The effectiveness of these improvements was evaluated on the set of open source projects containing 6 million lines of code.
This paper is devoted to a numerical simulation of 2D gas dynamics flows on uniform rectangular meshes using the Runge-Kutta-Discontinuous-Galerkin (RKDG) method. The RKDG algorithm was implemented with in-house C++ c...
详细信息
Code search is an important and well-studied task, but it usually means searching for code by a text query. We argue that using a code snippet (and possibly an error traceback) as a query while looking for bugfixing i...
详细信息
mpC is a medium level programming language for distributed memory machines (DMM). The language is an ANSI C superset based on the notion of a network comprising virtual processors of different types and performances c...
详细信息
mpC is a medium level programming language for distributed memory machines (DMM). The language is an ANSI C superset based on the notion of a network comprising virtual processors of different types and performances connected with links of different bandwidths. It allows the user to describe a network topology, create and discard networks, and distribute data and computations over the networks. In other words, the user can specify (dynamically) the topology of his application, and the mpC programming environment will use this (topological) information in run time to ensure the efficient execution of the application on any particular DMM. The paper outlines the principal features of mpC and its programming environment which make them suitable tools to write efficient and portable parallel programs for heterogenous DMM.
Finite state machines (FSM) are widely used in testing and verification of discrete event and hybrid systems. In order to guarantee fault coverage, various fault models and formal methods are utilized ranging from cla...
详细信息
ISBN:
(数字)9781665466059
ISBN:
(纸本)9781665466066
Finite state machines (FSM) are widely used in testing and verification of discrete event and hybrid systems. In order to guarantee fault coverage, various fault models and formal methods are utilized ranging from classical finite state machines (FSM) to logic circuits. The main problem for deriving an FSM-based test suite is the size of an FSM for real digital circuits, and thus, test engineers prefer to randomly derive a prefix of a test sequence and later on extend it with tests for hard detectable faults of a corresponding logic circuit. Usually at the latter step test sequences detecting hard detectable stuck-at faults are considered. In this paper, we are concerned about single functional faults, namely, $\boldsymbol{a},\boldsymbol{b}-\mathbf{faults}$ , of a system of Boolean functions of the combinational part of a sequential circuit such as deleting a variable or a cube in a SoP which represents the ON-set or the OFF-set of a corresponding Boolean function and experimentally evaluate fault coverage of a corresponding complete test suite with respect to fault coverage of tests returned by other methods. The experimental results show high fault coverage of test suites complete with respect to $\boldsymbol{a},\boldsymbol{b}-\mathbf{faults}$ and thus, it can be recommended to utilize such a test suite instead of a complete FSM-based test suite or to extend randomly generated test sequences for detecting $\boldsymbol{a},\boldsymbol{b}-\mathbf{faults}$ not killed with the prefix.
Static analysis is a general name for various methods of program examination without actually executing it. In particular, it is widely used to discover errors and vulnerabilities in software. Taint analysis usually d...
详细信息
ISBN:
(纸本)9781665423311
Static analysis is a general name for various methods of program examination without actually executing it. In particular, it is widely used to discover errors and vulnerabilities in software. Taint analysis usually denotes the process of checking the flow of user-provided data in the program in order to find potential vulnerabilities. It can be performed either statically or dynamically. In the paper we evaluate several improvements for the static taint analyzer Irbis [1], which is based on a special case of interprocedural graph reachability problem - the so-called IFDS problem, originally proposed by Reps et al. [2]. The analyzer is currently being developed at the Ivannikov institute for system programming of the Russian Academy of Sciences (ISP RAS). The evaluation is based on several real projects with known vulnerabilities and a subset of the Juliet Test Suite for C/C++ [3]. The chosen subset consists of more than 5 thousand tests for 11 different CWEs.
暂无评论