dziribot: rag based intelligent conversational agent for algerian arabic dialect

TL;DR

DziriBOT tackles the challenge of Algerian Darja by combining a dialect-focused NLU (DziriBERT) with Retrieval-Augmented Generation to ground responses in enterprise documentation. The system uses a five-tier pipeline with dual-script normalization, diverse embeddings, and a hybrid routing mechanism to cover 69 intents and open-domain queries. Empirical results show DziriBERT outperforming baselines in both Arabic-script and Arabizi inputs, with accuracy up to 92% on Latin-script data, while RAG improves scalability and factual grounding despite higher latency. The work demonstrates practical dialect-aware automation for Maghrebi telecoms and outlines a path toward real-time deployment through hardware acceleration and future few-shot and cross-dialect transfer learning.

Abstract

The rapid digitalization of customer service has intensified the demand for conversational agents capable of providing accurate and natural interactions. In the Algerian context, this is complicated by the linguistic complexity of Darja, a dialect characterized by non-standardized orthography, extensive code-switching with French, and the simultaneous use of Arabic and Latin (Arabizi) scripts. This paper introduces DziriBOT, a hybrid intelligent conversational agent specifically engineered to overcome these challenges. We propose a multi-layered architecture that integrates specialized Natural Language Understanding (NLU) with Retrieval-Augmented Generation (RAG), allowing for both structured service flows and dynamic, knowledge-intensive responses grounded in curated enterprise documentation. To address the low-resource nature of Darja, we systematically evaluate three distinct approaches: a sparse-feature Rasa pipeline, classical machine learning baselines, and transformer-based fine-tuning. Our experimental results demonstrate that the fine-tuned DziriBERT model achieves state-of-the-art performance. These results significantly outperform traditional baselines, particularly in handling orthographic noise and rare intents. Ultimately, DziriBOT provides a robust, scalable solution that bridges the gap between formal language models and the linguistic realities of Algerian users, offering a blueprint for dialect-aware automation in the regional market.