Identifying Banking Transaction Descriptions via Support Vector Machine Short-Text Classification Based on a Specialized Labelled Corpus

Abstract

Short texts are omnipresent in real-time news, social network commentaries, etc. Traditional text representation methods have been successfully applied to self-contained documents of medium size. However, information in short texts is often insufficient, due, for example, to the use of mnemonics, which makes them hard to classify. Therefore, the particularities of specific domains must be exploited. In this article we describe a novel system that combines Natural Language Processing techniques with Machine Learning algorithms to classify banking transaction descriptions for personal finance management, a problem that was not previously considered in the literature. We trained and tested that system on a labelled dataset with real customer transactions that will be available to other researchers on request. Motivated by existing solutions in spam detection, we also propose a short text similarity detector to reduce training set size based on the Jaccard distance. Experimental results with a two-stage classifier combining this detector with a SVM indicate a high accuracy in comparison with alternative approaches, taking into account complexity and computing time. Finally, we present a use case with a personal finance application, CoinScrap, which is available at Google Play and App Store. INDEX TERMS Machine learning, natural language processing, banking, personal finance management.

Deep learning enhancing banking services: a hybrid transaction classification and cash flow prediction approach

“…With regard to transaction categorization in the banking industry, there are plenty of commercial applications based on ML classification approaches, focusing mainly at personal transaction categorization solutions. A related research comes from [ 10 ], where short texts were utilized towards a novel system that combines Natural Language Processing techniques with Machine Learning algorithms to classify banking transaction descriptions for usage in a Personal Finance Management (PFM) application. A labelled dataset with real customer transactions was utilized exploiting existing solutions in spam detection by proposing a short text similarity detector to reduce training set size.…”

SBM: A Smart Budget Manager in banking using machine learning, NLP, and NLU

“…In Reference 34, the authors use a standard machine learning approach combined with a feature enrichment using the Brønnøysund Registry and the Google Places API as external semantic resources for the classification of real bank transactions. In Reference 35, the authors propose a short‐text SVM bank transaction classification system using a combination of meta‐information and linguistic knowledge (by relying on specialized lexica). They also reduced the training information with a short text similarity detector based on the Jaccard distance.…”

An introduction of transaction session‐induced security scheme using blockchain technology: Understanding the features of Internet of Things–based financial security systems

“…In the session span, TS3‐BT scheme is used to evaluate the metrics in time complexity (s) of observation is 7.45%, and in transaction, 7.72% is less. According to Garcia‐Mendez et al (2020), this proposed security scheme for user information satisfies less time complexity and transaction failures by predicting the active and passive transactions at different intervals and their associated financial security methods. With blockchain‐stored information, the time complexity is less at different transaction intervals.…”