The availability of public services to meet the needs of each community is one of the most important aspects of a country. People who require government assistance submit requests for assistance through public assistance offices, however, the number of complaints and requests received by these offices has become disproportionate to the number of people processing them. To address this, the researchers proposed the development of an automated system that can assist employees in streamlining the workflow within public assistance offices, which can help enhance workplace efficiency and quicken the resolution of cases. The system uses the Multinomial Naive Bayes algorithm to extract relevant information from emails and classify them based on the training data, determining the appropriate agency for the email to be sent to. Emails received are initially processed and then stored in a database that can be easily edited and re-processed. During testing, the researchers conducted a trial using a testing dataset of 150 emails, which took three and a half minutes (3m30s) to classify and send emails to their appropriate agencies, which is a significant improvement compared to the manual system that is able to process 300-500 emails within a 3-day timeframe. The researchers surveyed 45 respondents comprising students, teachers, professionals, and public assistance officers to evaluate the application. The respondents rated the application an average of 4.6 out of 5.0 across the categories given by the developers which corresponds to a Satisfactory rating. For future studies, the researchers recommended for the program to accommodate more classifications, separate the nature of concern and agency in the classification, implement alternative algorithms, implement stronger security measures, and add filtering options.