The efforts of computer scientists提高网络安全，科学家在努力

龚勇

為了克服基于网络的系统的弱点，世界各地的计算机科学家一直在尝试开发先进的入侵检测系统，提高网络的安全性。

主题语境：科技与信息 篇幅：314词 建议用时：7分钟

1 Networkbasedtechnology has becomeincreasingly widespread， and it is now beingused by countless individuals， professionals， andbusinesses worldwide. Despite the technologysadvantages， most networkbasedsystems arehighly weak to malicious attacks. The consequencesof a malicious attack on networkbasedsystems can be extremely severe and devastating. For instance， an attack on a power utilitynetwork could leave millions of individuals and offices without electricity， while attacks onsocial media networks can lead to the disclosure of user information.

2 To overcome the weaknesses of network?based systems， computer scientists world?wide have been trying to develop advanced intrusion （入侵） detection systems （IDSs）that could help to identify malicious attacks， increasing a network s safety. In recentyears， machine learning （ML） algorithms （算法） have been found to be particularly promisingfor automatically detecting attacks and intrusions on a networks functioning.

3 A key step in the development and training of ML?based IDSs is the selection ofdata features that a model can rely on or focus on when making predictions. Ideally， byanalyzing large datasets， researchers should be able to identify the most suitable featuresfor solving a given task using ML tools， and this is also applicable to intrusion detection.

4 Researchers at Canadian University Dubai in the UAE have recently developed anew feature selection method that could enable the development of more effective MLbasedIDSs. This method， presented in a paper prepublishedon arXiv， was found to perform remarkably well when compared with other commonly employed feature selectiontechniques.“Our goal is to study feature selection in network traffic data with the aim ofdetecting potential attacks，” Firuz Kamalov， Sherif Moussa， Rita Zgheib and OmarMashaal， the researchers who carried out the study， wrote in their paper.“We considervarious existing feature selection methods and propose a new feature selection algorithm toidentify the most effective features in network traffic data.”

Reading Check

1. Whats the function of the two examples mentioned in paragraph 1？

A. To present the power of networkbased systems.

B. To show the weaknesses of networkbased systems.

C. To explain the variety of malicious attacks.

D. To illustrate the serious results of malicious attacks.

2. How do IDSs improve a networks safety？

A. By preventing malicious attacks.

B. By recognizing malicious attacks.

C. By imitating malicious attacks.

D. By devastating malicious attacks.

3. Whats the key to the development of MLbased IDSs？

A. The settlement of tasks.

B. The analysis of datasets.

C. The prediction of the model.

D. The choosing of data features.

Language Study

Ⅰ. Useful expressions

for instance 例如；比如

rely on 依靠；依赖

be able to do 能够做……

compare with 与……相比较

Ⅱ. Difficult sentences

1. To overcome the weaknesses of networkbasedsystems， computer scientists worldwidehave been trying to develop advanced intrusion detection systems （IDSs） that couldhelp to identify malicious attacks， increasing a networks safety. 为了克服基于网络的系统的弱点，世界各地的计算机科学家一直在尝试开发先进的入侵检测系统，这类系统可以帮助识别恶意攻击，提高网络的安全性。

【点石成金】该句的主干为computer scientists worldwide have been trying to developadvanced intrusion detection systems （IDSs）；that could help to identify...safety为that引导的定语从句，修饰先行词intrusion detection systems，且关系代词that在从句中作主语；To overcome...systems为不定式结构作目的状语。

2. A key step in the development and training of MLbasedIDSs is the selection ofdata features that a model can rely on or focus on when making predictions. 基于机器学习的入侵检测系统开发和训练的关键步骤是选择模型在进行预测时可以依赖或关注的数据特征。

【点石成金】该句的主干为A key step is the selection of data features；in thedevelopment and training of MLbasedIDSs 為介词结构作后置定语，修饰A key step；that a model can rely...focus on为that引导的定语从句，修饰先行词data features，且关系代词that在从句中作宾语；when making predictions为when引导的时间状语从句的省略形式，省略了主语a model和be动词is。