Legal Text Classification in to Practice Areas Using Machine Learning

IJCSEC Front Page

Machine learning can be applied for various classification purposes in language processing. Application of machine learning to the legal domain remains a relatively new task. With the increasing ubiquity of the internet, individuals are looking more to internet resources to find relevant attorneys and to obtain answers to their legal questions. In this seminar report, a survey on various classifiers that can be applied for this application is conducted. the steps taken to build a machine learning classifier that successfully classifies legal questions or text into the most relevant practice area is described. We have created 6 different general categories that legal questions fall into. Categorizing legal questions into the correct practice area has many useful applications such as facilitating improved realtime feedback, information retrieval, relevant lawyer recommendations, and responses to users asking questions on Q&A websites.

Keywords: Multiclass Classification, Machine learning, Legal text.


  1. Chang, Y., Diesner, J. and Carley, K.M., 2012. Toward Automated Definition Acquisition From Operations Law. IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews), 42(2), pp.223-232.
  2. Zahoor, F. and Bajwa, I.S., 2014, August. Automatic Extraction of Catchphrases from Software License Agreement. In Intelligent Human-Machine Systems and Cybernetics (IHMSC), 2014 Sixth International Conference on (Vol. 2, pp. 189-193). IEEE.
  3. Lao, B. and Jagadeesh, K., Classifying Legal Questions into Topic Areas Using Machine Learning.
  4. Goyal, R.D., 2007, November. Knowledge based neural network for text classification. In Granular Computing, 2007. GRC 2007. IEEE International Conference on (pp. 542-542). IEEE.
  5. M. Ikonomakis, S. Kotsiantis, V. Tampakas, Text Classification Using Machine Learning Techniques, Wseas Transactions on Computers, Vol, 4, Iss: 8, pp. 966-974, 2005.