TF-IDF vs Word Embeddings for Morbidity Identification in Clinical Notes: An Initial Study 1 2 Danilo Dess`ı [0000−0003−3843−3285], Rim Helaoui [0000−0001−6915−8920], Vivek 1 1 Kumar [0000−0003−3958−4704], Diego Reforgiato Recupero [0000−0001−8646−6183], and 1 Daniele Riboni [0000−0002−0695−2040] 1 University of Cagliari, Cagliari, Italy {danilo dessi, vivek.kumar, diego.reforgiato, riboni}@unica.it 2 Philips Research, Eindhoven, Netherlands
[email protected] Abstract. Today, we are seeing an ever-increasing number of clinical notes that contain clinical results, images, and textual descriptions of pa-tient’s health state. All these data can be analyzed and employed to cater novel services that can help people and domain experts with their com-mon healthcare tasks. However, many technologies such as Deep Learn-ing and tools like Word Embeddings have started to be investigated only recently, and many challenges remain open when it comes to healthcare domain applications. To address these challenges, we propose the use of Deep Learning and Word Embeddings for identifying sixteen morbidity types within textual descriptions of clinical records. For this purpose, we have used a Deep Learning model based on Bidirectional Long-Short Term Memory (LSTM) layers which can exploit state-of-the-art vector representations of data such as Word Embeddings. We have employed pre-trained Word Embeddings namely GloVe and Word2Vec, and our own Word Embeddings trained on the target domain. Furthermore, we have compared the performances of the deep learning approaches against the traditional tf-idf using Support Vector Machine and Multilayer per-ceptron (our baselines).