A Comparison of Text Classifiers on IT Incidents Using WEKA

Iscen, Erdal Sever; Gürbüz, Mustafa Zahid

doi:10.1109/ubmk.2019.8907063

Cited by 4 publications

(13 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nevertheless, a statement can be made about which ML algorithms and solutions have proven themselves during the past 5 years not in one paper only, but in various papers and hence in various use cases. Generally, ML algorithms like SVM and RF have proven to be more accurate and precise than "older" rules-based approaches or Decision Trees [2,8,23]. Also, there is some evidence that DNNs could outperform the present best-performing algorithms SVM and RF, especially in the case of large training data and many classes to classify tickets in [21,29].…”

Section: Discussionmentioning

confidence: 99%

“…In these 6 papers accuracies between 63% [37] and 98% [18] were reached. Those results are heavily depending on the data set used, but overall the papers an accuracy between 80-90% for SVM seems reasonable [1,2,8,9,18,37].…”

Section: Machine Learning Algorithms Usedmentioning

confidence: 94%

“…İşcen and Gürbüz [2] x Lo, Tiba [25] x x Mukunthan and Selvakumar [26] x Nayebi, Dicke [6] x x Mandal, Agarwal [11] x Misra and Podder [27] (x) x Palacios, Morillas [28] (x) x Shanmugalingam, Chandrasekara [29] x…”

Section: Sent Pred Othermentioning

confidence: 99%

“…According to Revina, Buza [23], the earliest approaches of support ticket classification were ruledbased approaches that were outperformed by RF and SVM. This probably is the reason that mostly in the years 2018 and 2019, RF and SVM were the approaches most implemented and evaluated and also best performing [2,6,8,9,30,36]. Since 2019, some authors also propose artificial-neuronal-networkbased solutions, because of the good performance of neuronal network approaches in other fields of ML [19,21,23,29].…”

Section: <60% Accuracymentioning

confidence: 99%

“…Providing technical support for own IT products is an integral part of software developing or software providing companies [1,2]. For this purpose, most companies provide their customers support ticket systems (STSs), in which users can create incident tickets describing their problem or request [1].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Improving Support Ticket Systems Using Machine Learning: A Literature Review

Fuchs¹,

Drieschner²,

Wittges³

2022

Proceedings of the Annual Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

Processing customer support requests via a support ticket system is a key-element for companies to provide support to their customers in an organized and professional way. However, distributing and processing such tickets is much work, increasing the cost for the support providing company and stretching the resolution time. The advancing potential of Machine Learning has led to the goal of automating those support ticket systems. Against this background, we conducted a Literature Review aiming at determining the present state-of-the-art technology in the field of automated support ticket systems. We provide an overview about present trends and topics discussed in this field. During the Literature Review, we found creating an automated incident management tool being the majority topic in the field followed by request escalation and customer sentiment prediction and identified Random Forrest and Support Vector Machine as best performing algorithms for classification in the field.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Machine Learning Algorithms Usedmentioning

confidence: 94%

Section: Sent Pred Othermentioning

confidence: 99%

Section: <60% Accuracymentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Improving Support Ticket Systems Using Machine Learning: A Literature Review

Fuchs¹,

Drieschner²,

Wittges³

2022

Proceedings of the Annual Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

show abstract

Knowledge-based Intelligent System for IT Incident DevOps

Ahmed

Singh

Doherty

et al. 2023

2023 IEEE/ACM International Workshop on Cloud Intelligence &Amp; AIOps (AIOps)

View full text Add to dashboard Cite

The automation of IT incident management (i.e., handling of any unusual events that hamper the quality of IT services) is a main focus in Artificial Intelligence for IT Operations (AIOPS). The success and reputation of large-scale firms depend on their customer service and helpdesk system. These systems tend to handle client requests and track customer service agent interactions. In this research, we present a complete knowledge-based system that automates two core components of IT incident service management (ITSM): (1) Ticket Assignment Group(TAG) and (2) Incident Resolution (IR). Our proposed system bypasses the 4 core steps of the traditional ITSM process, including data investigation, event correlation, situation room collaboration, and probable root cause. It provides immediate solutions that can save companies key performance indicator(KPIs) resources and reduce the mean time to resolution (MTTR). The experiment used an industrial, real-time ITSM dataset from a prominent IT organization comprising 500,000 real-time incident descriptions with encoded labels. Furthermore, our systems are then evaluated with an open-source dataset. Compared to the existing benchmark methodologies, there is a 5% improvement in terms of Accuracy score. The study demonstrates AI automation capabilities in incident handling (TAG and IR) for large realworld IT systems.

show abstract

Multiple Severity-Level Classifications for IT Incident Risk Prediction

Ahmed¹,

Singh²,

Doherty³

et al. 2022

2022 9th International Conference on Soft Computing &Amp; Machine Intelligence (ISCMI)

View full text Add to dashboard Cite

The adoption of Artificial Intelligence (AI) is now widespread in Information Technology (IT) support. A particular area of interest is in the automation of IT incident management (i.e., the handling of an unusual event that hampers the quality of IT services in the most optimized manner). In this paper, we propose a framework using state-of-art algorithms to classify and predict the severity of such incidents (commonly labeled as High, Medium, and Low severity). We argue that the proposed framework would accelerate the process of handling IT incidents with improved accuracy. The experimentation was performed on the IT Service Management (ITSM) dataset containing 500,000 real-time incident descriptions with their encoded labels (Dataset 1) from a reputable IT firm. Our results showed that the Transformer models outperformed machine learning (ML) and other deep learning (DL) models with a 98% AUC score to predict the three severity classes. We tested our framework with an open-access dataset (Dataset 2) to further validate our findings. Our framework produced a 44% improvement in AUC score compared to the existing benchmark approaches. The results show the plausibility of AI algorithms in automating the prioritization of incident processing in large IT systems.

show abstract

A Comparison of Text Classifiers on IT Incidents Using WEKA

Cited by 4 publications

References 4 publications

Improving Support Ticket Systems Using Machine Learning: A Literature Review

Improving Support Ticket Systems Using Machine Learning: A Literature Review

Knowledge-based Intelligent System for IT Incident DevOps

Multiple Severity-Level Classifications for IT Incident Risk Prediction

Contact Info

Product

Resources

About