Oleh Diposting pada

Urang biasa ngaran ramatloka ditungtungan ku .com, .org, .net, jeung saterusna. Sababaraha taun ka tukang geus katempo ekstensi domain anyar muncul, kayaning .aero, .club jeung sajabana. Ieu katelah domain tingkat luhur (TLDss), sarta daptar, geus lila, boga tambahan anyar occasional. Google ngumumkeun dina Méi yén dalapan domain deui éta sadia, dua di antarana éta ngalelep teu bisa dibédakeun tina ekstensi file populér: .zip jeung .mov. Léngkah ieu dikritik ku para ahli IT sareng infosec, sabab ngan ukur ngajamin kabingungan, penanganan tautan anu pabalatak, sareng pola phishing énggal.

Kumaha galau .zip na .zip

File ZIP sareng MOV parantos aya mangpuluh-puluh taun: .zip mangrupikeun standar arsip de facto, sareng .mov mangrupikeun salah sahiji wadah pidéo anu paling populér. Google ngarahkeun domain MOV sareng ZIP anyar ieu ka teknisi, tapi saleresna sayogi pikeun saha waé sareng pikeun tujuan naon waé.

Ayeuna, ngan ukur kontéks anu tiasa ngabantosan anjeun terang upami ZIP atanapi MOV mangrupikeun halaman wéb atanapi file nalika anjeun mendakan, ucapkeun, update.zip. Nanging, kontéks mangrupikeun hal anu tiasa kahartos ku manusa, sanés komputer, ku kituna rujukan sapertos kitu tiasa nyababkeun masalah dina sagala jinis aplikasi, sapertos Twitter:

Tweets mentioning .zip jeung .mov. file

Tweet nu jelas nujul kana file, tapi Twitter renames file ka link web. Upami aya anu ngadaptarkeun domain test.zip sareng movie.mov, anu ngaklik tautan file tiasa janten korban skéma phishing anu tangtu.

Manggihan salapan bédana.

Panaliti kaamanan mr.d0x mendakan cara anu sanés pikeun ngamangpaatkeun domain .zip pikeun phishing. Téhnik anu dijelaskeun ku anjeunna, anu disebat file-archiver-in-the-browser, ngalibatkeun ngagunakeun situs anu meniru antarmuka utilitas arsip. Pamaké, percanten yén anjeunna muka file .zip, sabenerna dialihkeun ka situs nu ngaranna sarua jeung tinimbang daptar file anjeunna nilik URL nu bisa ngakibatkeun mana. Salaku conto, aranjeunna tiasa nyumputkeun tautan pikeun ngaunduh executable malware, atanapi naroskeun kredensial damel pikeun ngaksés sababaraha dokumén. Dokumén anu sami ogé ngajelaskeun mékanisme ngirim anu pikaresepeun nganggo Windows File Explorer. Upami panyerang tiasa ngayakinkeun korbanna pikeun milarian file .zip anu teu aya, File Explorer bakal otomatis muka situs dina domain anu nami anu sami.

Ancaman phishing nyata, sareng sababaraha .zip situs phishing eksploitasi téma apdet Windows parantos katingal.

Henteu yén ieu pertama kalina urang ningali kabingungan anu sami sareng ieu. Salah sahiji domain aslina, .com, oge extension légal pikeun executables aktip dipaké dina MS-DOS (jeung versi saméméhna tina Windows), bari extension .sh dipaké pikeun Aksara Unix idéntik jeung TLD pikeun Téritori peuntas laut Inggris di. St Héléna, Ascension jeung Tristan da Cunha. Masih, ZIP sareng MOV, anu populer di kalangan pamiarsa anu kirang téknis, nyababkeun masalah poténsial pikeun pangguna sareng pangurus sistem sami. Sanaos anjeun hilap phishing sakedap, kaayaan sapertos anu dijelaskeun dina tweet di luhur tiasa lumangsung dina seueur aplikasi anu otomatis ngolah téks sareng nyorot tautan. Ku alatan éta, iraha waé téks anu ngandung nami file tiasa dirobih janten téks anu ngandung hyperlink ka situs wéb éksternal. Skéma phishing atanapi henteu, ieu tiasa nyababkeun sahenteuna sababaraha kasulitan upami henteu kabingungan. Didatangan financial reports.zip ningali sorangan.

Tips pikeun pamaké

Munculna ZIP sareng domain MOV henteu bakal nyababkeun parobihan drastis dina ékosistem phishing sareng scam online – éta ngan ukur bakal nambihan hiji deui senjata kana arsenal hacker anu parantos masif. Ku alatan éta, tips anti phishing dawam kami tetep unchanged: study link mana wae taliti saméméh ngaklik; Waspada kantétan sareng URL dina email anu henteu diperyogikeun; ulah klik link curiga; sareng pastikeun ngagunakeun kaamanan anu leres dina sadaya alat anjeun – bahkan smartphone sareng Mac.

Tips kanggo pangurus

Sababaraha pamaké sigana teu malire saran di luhur, jadi gumantung kana kumaha organisasi Anjeun beroperasi, Anjeun bisa jadi kudu nyetel aturan kaamanan misah pikeun ngaran domain .zip jeung .mov. Léngkah-léngkah anu mungkin kalebet panyeken tautan anu langkung ketat atanapi bahkan ngahalangan pangguna tina ngadatangan situs wéb dina domain ieu dina komputer perusahaan. Ieu sanés tanpa precedent: domain .bit seueur diblokir sareng laun-laun maot kusabab panyaluran tautan jahat dina 2018-2019.

Kadatangan domain ZIP sareng MOV mangrupikeun alesan anu saé pikeun ngalakukeun – atanapi malikan deui! – pelatihan infosec pikeun pagawé (kalayan fokus kana deteksi phishing).

Kami ngarékoméndasikeun yén pangurus IT nguji sistem bisnis utama naon waé anu ngolah tautan pikeun ningali kumaha nanganan situs wéb .zip sareng .mov, sareng upami pamakean file ZIP dibarengan ku épék anu teu dihoyongkeun. Sistem email, aplikasi olahtalatah instan perusahaan, sareng ladenan babagi file karyawan kedah diawaskeun sacara saksama, sabab ieu mangrupikeun kabingungan anu paling dipikaresep. Fitur nu teu dihoyongkeun, kayaning wangunan link otomatis dumasar kana pola ngaran nu tangtu, bisa ditumpurkeun pikeun ZIP jeung MOV atawa sakabéhna.


#Résiko #kaamanan #tina #domain #.zip #sareng #.mov

Oleh Diposting pada

The detailed calculation of the three measurements will be presented in section III. In concentration in the plan, we initially present the information assortment technique that contains the development of data sets used for the location of malware and family identification, and the extraction of highlights from the APK records. Then we detail the instincts of the three proposed quantitative measurements, as well as their calculation conditions. Construction of the data set: Five widely used malware data sets are evaluated in our work. 111 The Single AMD data set contains 24,650 examples, which is a tedious task to solve the clarification results for all examples.

In this way, we arbitrarily select a fifth examples of each family. For accommodation, they are named as DATASET-I, DATASET-II, DATASET-III, DATASET-IV and DATASET-V. Its representations are recorded in Table II, where segments 2-3 stops the number of families and the amount of malware tests, sections 4-6 summary the largest, lowest and normal size of the tests, and the section 7 Record the delivery season of the season of the delivery season of the delivery season of the delivery season of the delivery season of the delivery season of the delivery season of the delivery season of Testing comparison.

In addition, the results are phenomenal taking into account the short preparation length and a hyper-united rationalization was not carried out. To represent the efficient element of movement learning, we directed a similar essay that uses the convolution network shown in Fig. 2. To contrast and the prepredible networks, the preparation was ended for 25 ages using equivalent hyper-boundariales. The acquired accuracy does not reflect the precision of the prognosis of individual malware families. To that end, we have prosecuted the disorder network shown in Fig. 5 for Resnet152. All others are comparable. Specifically, the low accuracy of the Simda forecast is generally due to the small number of tests. In this work we examine the viability of movement learning for malware group.

To that end, we have made probes four networks prepared to characterize malware. Specifically, the group was carried out in the data set challenged by the Microsoft classification that was completely changed to gray -scale images. All designs in the organization gave more than 95% precision using not many ages of preparation. This is exceptionally encouraging since they were prepared in Imagemet. This shows that movement learning is solid since each of the various organizations gave a similar way of behaving. Another point of view that is worth examining that we transmit to future work, is the deduction season of network models concentrated on small PC devices in chip such as Nvidia Jetson Nano.

It offers your panda without risk DVD, the starting rescue circle and the transport vaccine of the Universal Panda series to keep your units safe. You will defend yourself concentrated, which is not restricted by a singular PC, only in the light of the fact that this technique is fog in view of the web. There is a month-to-month membership support known as Panda Anti-Infection Professional next year, which therefore only spends $ 40 every year, and also 3 of one computers will probably be safeguard. Fundamental security is indispensable regarding the web, and also penetration only by malware and also the malware is undoubtedly typical at this time.

You must safeguard your PC through the Trojans and also malware, explicitly essentially in the light of the fact that they could close their PC for their current character in a cut jump. Its UGG boots and, in addition, essential security products, including next year, of which we have currently explored in this publication could help you close these events. We really trust that you have incredible advertising with UGG Boots Olet. Its best classifier has a good performance with a 0.98 F1 score. In addition, they survey their location model in a large data set of more than 27,000 vindictive applications. During the composition of this document, different creators used the Sherlock data set, but to a significantly more restricted degree than this exploration.

7 classifiers were prepared for malware discovery: isotonic regression, random forest, decision trees, gradient trees, multicapa perceptron, SVM and logistics regression. The information was adjusted so that half of the brands were harmless and half were pernicious. Only information from the third quarter of 2016 was considered, which implies that three main types of malware (Madware, ransomware and click-jacking) were incorporated for preparation and tests, with CPU, network traffic, drums and outstanding aspects of the cycle . Increased gradient trees had the best results with a 0.91% F1 score and a FPR of 0.09%. Its random forest classifier also worked similarly. These obtain improved results than the previous work, which could be due to the default number of types of malware included.