International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 13 Issue: 01 | Jan 2026 www.irjet.net p-ISSN: 2395-0072
Case-Based Reasoning for Mobile Helpdesk System
Dnyaneshwari Nitin Patil¹, Achal Ekanath Hujare², Pranita Shrikant Dalavi³, Archana. V. Jadhav4
1,2,3,4 Student, Computer Engineering, Dr. D. Y. Patil Polytechnic, Kolhapur, India
Abstract - In the modern digital era, mobile users expect fast and dependable technical support. Traditional helpdesk systems based on FAQs and manual assistance are often slow and inefficient. This paper presents an intelligent mobile helpdesk system using Case-Based Reasoning (CBR), where user problems are solved by reusing solutions from previously resolved cases. When a relevant case is found, the system delivers an immediate solution; otherwise, the issue is forwarded to an administrator and stored for future learning. The proposed system reduces response time, decreases manual effort, and improves overall user satisfaction, making it an effective and practical helpdesk solution.
Key Words: Case-Based Reasoning (CBR), NLP, Mobile Helpdesk System, Intelligent Support System, Knowledge-Based System, Automated Problem Solving
1.INTRODUCTION
The rapid growth of mobile technology has increased the demand for fast and reliable technical support. Mobile users frequently face issues related to applications, software updates, and network connectivity. Traditional helpdesk systems based on FAQs and manual assistance are often slow, inefficient, and unable to handle repeated or similar queries effectively. To overcome these limitations, intelligent helpdesk systems are required. Case-Based Reasoning (CBR) is an artificial intelligence technique that solves new problems by reusing solutions from previously solved cases. By learning from past experiences, CBR systems provide quick and consistent responses. This paper proposes an Intelligent Mobile HelpdeskSystemusingCase-BasedReasoning.Thesystem retrieves solutions from a case database and provides instant support to users. Unresolved queries are handled by an administrator and stored for future learning. The proposed approach reduces response time, minimizes manualeffort,andimprovesoverallhelpdeskefficiency
2.LITERATURE SURVEY
Introduces a system that combines large language models (LLMs) with retrieval-augmented generation to improve case-based reasoning. It can handle fuzzy or imprecise descriptions of cases, making it more flexible. Useful for medical/healthcareandlegaldomains;couldbeadaptedto helpdesk + healthcare. [1] This paper shows how you can use fuzzy logic plus adaptive similarity weighting in a
diagnosissystemwithCBR.Themethodsherearegoodfor retrieval & adaptation of past similar cases, which is exactly useful if your chat bot needs to pick a close matching previous case when no exact answer exists. [2] This one is directly about applying CBR in helpdesk systems. It studies processes by which helpdesk systems can retrieve past similar cases to resolve new tickets, reducing resolution time. Good match with your project idea. [3] This describes IHDF, a system used in a bank’s helpdesk environment for network/computer fault management. It uses hybrid knowledge representation, fuzzy/neighbor similarity matching, indexing, etc. Very practicalcaseofhowtobuildandmaintainacasebaseand perform retrieval and reuse. [4] Introduces a reasoning system that starts with a simple/coarse set of features, then refines to more detailed features if necessary. So when matching a new query, system tries simpler comparisons first, and if not good enough, goes deeper. This is helpful when performance matters (mobile environment) and helps reduce computation. [5] Classic foundational paper on what CBR is, how it works (storing cases, retrieving, adapting, retaining), memory/organization/indexing, etc. Good for theory/backgroundinyourdocumentation.[6]Aamodtand Plaza presented the classical Case-Based Reasoning cycle consisting of retrieve, reuse, revise, and retain phases. Their work emphasizes continuous learning and knowledge reuse, which forms the foundation of modern CBR-based intelligent systems, including helpdesk applications. [7] Pal and Shiu proposed an adaptive CaseBased Reasoning system that dynamically updates similarity measures based on new cases. This approach improves long-term system accuracy and learning efficiency, making it suitable for intelligent support systems. [8] Lu et al. introduced a coarse-to-fine CaseBasedReasoningapproachwherethesysteminitiallyuses simple features and gradually applies detailedfeaturesfor accurate matching. This method improves performance and reduces computational cost, especially in mobile environments. [9] Watson discussed real-world industrial applicationsofCase-BasedReasoningincustomersupport and troubleshooting systems. The study highlights how CBR reduces resolution time and improves decision consistencyinhelpdeskenvironments.[10]
3. PROPOSED SYSTEM
The proposed system is an intelligent mobile helpdesk designed using Case-Based Reasoning (CBR) to provide fast and accurate technical support. Users submit their
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 13 Issue: 01 | Jan 2026 www.irjet.net p-ISSN: 2395-0072
queries through a mobile application, which are first processed using Natural Language Processing (NLP) to clean and standardize the input. The processed query is then handled by the CBR engine, which retrieves similar previously solved cases from the case database. The most relevant solution is reused and delivered instantly to the user.Ifnosuitablecaseisfound,thequeryisforwardedto the administrator for resolution. Once resolved, the new problemandsolutionarestoredinthedatabaseforfuture learning. The system also supports auto-suggestions and real-time notifications to enhance user experience. By continuously learning from new cases, the proposed system reduces response time, minimizes manual effort, andimprovesoverallhelpdeskefficiency.
Case-based reasoning (CBR)
Retrieve: Search for similar past cases from the casedatabase.
Reuse: Apply and provide the best-matched solutiontotheuser.
Revise: If required, forward the query to the administratorforcorrection.
Retain: Storethenewcase anditssolutioninthe databaseforfutureuse.
Retain(save)
Problem NewCase Retrieved case Reused case Learned case Case Base
Retrieve(find)
Reuse
SimilarCase Handling Limited Efficientretrievalof similarcases
UserInteraction Webormanual interface Mobileapplication interface
Accuracy Moderate High
AdminEffort High Low
Scalability Limited High
Automation Level Partial High
5.ADVANTAGES
•Providesfasterresponsebyreusingpreviouslysolved cases
•Reducesworkloadanddependencyonhumanhelpdesk staff
•Self-learningsystemthatimprovesaccuracyovertime
•Ensuresconsistentandreliablesolutionsforsimilar queries
•Handleslargenumbersofuserrequestsefficiently
•Scalableandsuitableformobile-basedenvironments
•Improvesoverallusersatisfactionandsupportquality
6.CONCLUSION
Repaired/ testedcase
Revise(control)
Fig:Case-BasedReasoning(CBR)
4. COMPARATIVE ANALYSIS
Parameter Existing System Proposed System
SystemType Traditional/Semiautomatedhelpdesk IntelligentCBRbasedmobile helpdesk
QueryHandling Keywordormanual based NLP-based intelligentquery processing
ResponseTime Slowtomoderate Fastandreal-time
Learning Capability Noself-learning Self-learningusing CBRcycle
This paper presented an intelligent mobile helpdesk system based on Case-Based Reasoning (CBR) to address the limitations of traditional helpdesk solutions. Conventional systems relying on FAQs and manual support often suffer from slow response times, high operational costs, and limited adaptability. The proposed system overcomes these issues by reusing solutions from previously resolved cases, enabling fast and accurate problem resolution. By following the standard CBR cycle of retrieve, reuse, revise, and retain, the system continuously learns from new queries and improves its performance over time. The integration of Natural Language Processing further enhances query understanding and similarity matching. Administrative intervention ensures correctness for unresolved queries, while case retention strengthens the knowledge base. Overall, the proposed system reduces manual workload, improvesresponseefficiency,andprovidesconsistentand reliable support to users. Its scalable and self-learning nature makes it suitable for modern mobile-based helpdesk environments and lays a strong foundation for
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 13 Issue: 01 | Jan 2026 www.irjet.net p-ISSN: 2395-0072
futureenhancementsusingadvancedartificialintelligence techniques.
7.REFERENCES
[1] CaseGPT: a case reasoning framework based on language models and retrieval-augmented generation RuiYang,2024.
[2] A Logic and Adaptive Approach for Efficient Diagnosis SystemsusingCBR IbrahimElBitaretal.,2012.
[3] Application of Case-Based Reasoning in Helpdesk Systems Riyan Apriyanto, Ika Intan Rahmawati, etc., 2019(ICICConference).
[4] An integrated case-based reasoning approach for intelligenthelpdeskfaultmanagement (IntelligentHelp DeskFacilitator,Singapore)
[5] Case-Based FCTF (From Coarse to Fine) Reasoning System JingLuetal.,MDPI,2015.
[6] Case-Based Reasoning: A Research Paradigm StephenSlade,AIMagazine,1991.
[7] A. Aamodt and E. Plaza, “Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approaches,” AICommunications,vol. 7, no.1, pp. 39–59,1994.
[8] S. K. Pal and S. C. K. Shiu, Foundations of Soft CaseBasedReasoning,Wiley,2004.
[9] J. Lu, D. Ruan, and G. Zhang, “A Coarse-to-Fine CaseBased Reasoning Approach,” Knowledge-Based Systems, vol.28,pp.21–34,2015.
[10] A. Watson, “Applying Case-Based Reasoning: Techniques for Enterprise Systems,” Morgan Kaufmann, 1999