Complaint Analysis and Classification for Economic and Food Safety

Abstract

Governmental institutions are employing artificial intelligence techniques to deal with their specific problems and exploit their huge amounts of both structured and unstructured information. In particular, natural language processing and machine learning techniques are being used to process citizen feedback. In this paper, we report on the use of such techniques for analyzing and classifying complaints, in the context of the Portuguese Economic and Food Safety Authority. Grounded in its operational process, we address three different classification problems: target economic activity, implied infraction severity level, and institutional competence. We show promising results obtained using feature-based approaches and traditional classifiers, with accuracy scores above 70%, and analyze the shortcomings of our current results and avenues for further improvement, taking into account the intended use of our classifiers in helping human officers to cope with thousands of yearly complaints.

Supporting Complaints Investigation for Nursing and Midwifery Regulatory Agencies

“…Decision Support Systems. Many NLP systems have been developed to process text data (such as records, reports, scientific papers, and social media posts) to assist in making highly critical decisions, in domains like healthcare (Bampa and Dalianis, 2020;Mascio et al, 2020;Feng et al, 2020;Proux et al, 2009), finance (Kogan et al, 2009;Wang et al, 2013), business and management (Dong and Wang, 2015;Assawinjaipetch et al, 2016;Filgueiras et al, 2019), and legislation (Rabelo et al, 2019;Soh et al, 2019;Shaffer and Mayhew, 2019). Our work proposes the first decision support system to process nursing/midwifery complaints.…”

“…Hence, large neural network models do not always outperform classic featurerich models and careful model selection is often necessary. For example, Filgueiras et al (2019) found that, in an economic activity classification task, the SVM (Cortes and Vapnik, 1995) with TF-IDF (Salton and Buckley, 1988) representations performed better than an LSTM network (Hochreiter and Schmidhuber, 1997). On the other hand, Assawinjaipetch et al (2016) and Mullenbach et al (2018) showed that in complaint and clinical classification tasks, RNNs (Cho et al, 2014) or CNNs (Kim, 2014) with pre-trained word2vec embeddings (Mikolov et al, 2013) outperformed the classic machine learning models with bag-of-words representations.…”

Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing: System Demonstrations

“…Decision Support Systems. Many NLP systems have been developed to process text data (such as records, reports, scientific papers, and social media posts) to assist in making highly critical decisions, in domains like healthcare (Bampa and Dalianis, 2020;Mascio et al, 2020;Proux et al, 2009), finance (Kogan et al, 2009;Wang et al, 2013), business and management (Dong and Wang, 2015;Assawinjaipetch et al, 2016;Filgueiras et al, 2019), and legislation (Rabelo et al, 2019;Soh et al, 2019;Shaffer and Mayhew, 2019). Our work proposes the first decision support system to process nursing/midwifery complaints.…”

“…Hence, large neural network models do not always outperform classic featurerich models and careful model selection is often necessary. For example, Filgueiras et al (2019) found that, in an economic activity classification task, the SVM (Cortes and Vapnik, 1995) with TF-IDF (Salton and Buckley, 1988) representations performed better than an LSTM network (Hochreiter and Schmidhuber, 1997). On the other hand, Assawinjaipetch et al (2016) and Mullenbach et al (2018) showed that in complaint and clinical classification tasks, RNNs or CNNs with pre-trained word2vec embeddings (Mikolov et al, 2013) outperformed the classic machine learning models with bag-of-words representations.…”

Fine-Grained Element Identification in Complaint Text of Internet Fraud

“…Different from existing complaint-related work [6,7,21] that regards every single complaint text as independent, our corpus contains a hierarchical relationship between clauses and paragraphs. Therefore, except for categorical statistics, we further explore the distribution and relation of clauses in specific paragraph.…”