Explore tweets tagged as #vulnerabilityprediction
Thrilled to meet Pamela Phan, Deputy Assistant Secretary for Asia at U.S. Department of Commerce. Had fruitful conversation about Securin's products and #vulnerabilityresearch and US's mission to secure the cyber landscape. #cybersecurity #vulnerabilityprediction #ransomware
0
2
7
Excited to have shared Securin's research on #VulnerabilityPrediction and #Ransomware with US Ambassador to Singapore @JonathanKaplan with @GoIvanti . Committed to defending the homeland and contributing to global security. #cybersecurity #CyberDefense #StopRansomware
0
1
3
Excited to have shared Securin's research on #VulnerabilityPrediction and #Ransomware with US Ambassador to Singapore @JonathanKaplan along with our partner @GoIvanti . Committed to defending the homeland and contributing to global security. #cybersecurity #USmission
0
9
7
Are you interested in learning how #DeepLearning could potentially improve the accuracy of #VulnerabilityPrediction models in cross-project vulnerability prediction? Have a look at our latest paper! .Pre-print: 👉
1
0
3
🧵(5/5) 5⃣ Finally, there is a need for better #DataSharing, #Governance & #QualityControl of #vulnerability #datasets, to promote future work, increase the reliability of produced outcomes, & advance this field of research. #SLR.#VulnerabilityPrediction.#DataPreparation.
🧵(4/5) 4⃣ Current #data #solutions aim to address #challenges in data #accessability, #effort, & #noise, but they are imperfect solutions. Many potentially important #data #quality attributes are yet unconsidered by researchers, such as data #understandability & #timeliness.
0
2
6
🧵(2/5) 2⃣ Existing #vulnerability #datasets are assumed to lack #generalizability. Synthetic datasets are not considered to be representative of real-world code, & most #datasets are limited to a small set of application contexts. #SLR.#VulnerabilityPrediction.#DataQuality.
🧵of key contributions of our recently accepted #TSE paper👇. (1/5) 1⃣ Presents a comprehensive source of information regarding SVP data preparation #Practices #Challenges & #Solutions, including a #Taxonomy of SV data challenges -> useful to guide future SVP #Research & Practice.
0
1
5
🧵(3/5) 3⃣ #Label #correctness is a significant issue for many #vulnerability #datasets due to incomplete reporting & #localization issues stemming from the original data sources. It is also highly error-prone. #SLR.#VulnerabilityPrediction.#DataQuality.#DataPreparation.
🧵(2/5) 2⃣ Existing #vulnerability #datasets are assumed to lack #generalizability. Synthetic datasets are not considered to be representative of real-world code, & most #datasets are limited to a small set of application contexts. #SLR.#VulnerabilityPrediction.#DataQuality.
0
1
6
Check out our recently accepted paper at #MSR2022 that investigates #NoisyLabel learning methods applied to #Software #VulnerabilityPrediction 👇 #CREST_Research @CSCRCoz.
Using #MachineLearning 4 #software #security & troubled by noisy #data? @crest team provides an #innovative solution to #SVP using #Noisy Label Learning; accepted in @msrconf, funded by @CSCRCoz & led by @CroftRoland involving @Huaming_Chen & me; #GameOn
0
1
8
🔔🔔🔔 #MDPIfutureinternet [New Published Papers in 2023] . Title: #QuantumMachineLearning for Security Assessment in the Internet of Medical Things (#IoMT). #vulnerabilityprediction #InternetofThings #InternetofMedicalThings. @ComSciMath_Mdpi.
0
0
1
In IoTAC, we built #vulnerabilityprediction models that were based on traditional #textmining techniques and #ML learning algorithms. The @DOSSprojectHE will extend our prior work with AI technologies, focusing on Large Language Models (LLMs) that have emerged recently.
In the latest DOSS Insight by @siavvasm at @CERTHellas you can find out how the #LLM-based #vulnerability Prediction Mechanism will constitute a core element of the Component Tester module in the DOSS project. #HorizonEU #softwaresecurity #iotsecurity.
0
0
1
Significant attention should be paid 2 #DataQuality b4 developing #AI models 4 #VulnerabilityPrediction 2 achieve its effectiveness in practice. Our @crest_uofa research provided several key recommendations 2 guide researchers in #DataPreparation.
🧵(3/5) 3⃣ #Label #correctness is a significant issue for many #vulnerability #datasets due to incomplete reporting & #localization issues stemming from the original data sources. It is also highly error-prone. #SLR.#VulnerabilityPrediction.#DataQuality.#DataPreparation.
0
4
6