Search

Natural Language Understanding is one of the essential tasks for building clinical text-based applications. Understanding of these clinical texts can be achieved through Vector Space Models and Sequential Modelling tasks. This paper is focused on sequential modelling i.e. Named Entity Recognition and Part of Speech Tagging by attaining a state of the art performance of 93.8% as F1 score for i2b2 clinical corpus and achieves 97.29% as F1 score for GENIA corpus. This paper also states the performance of feature fusion by integrating word embedding, feature embedding and character embedding for sequential modelling tasks. We also propose a framework based on a sequential modelling architecture, named MedNLU, which has the capability of performing Part of Speech Tagging, Chunking, and Entity Recognition on clinical texts. The sequence modeler in MedNLU is an integrated framework of Convolutional Neural Network, Conditional Random Fields and Bi-directional Long-Short Term Memory network.

Understanding the meaning of protein sequences is tedious with human efforts alone. Through this work, we experiment an NLP technique to extract features and give appropriate representation for the protein sequences. In this paper, we have used GloVe representation for the same. A dataset named Swiss-Prot has been incorporated into this work. We were able to create a representation that has comparable ability to understand the semantics of protein sequences compared to the existing ones. We have analyzed the performance of representation by the classification of different protein families in the Swiss-Prot dataset using machine learning technique. The analysis done by us proved the significance of this representation.

Many healthcare-related applications use pretrained embeddings, but these are often trained over general corpus which is mostly downstreamed to certain particular application. One problem noticed among such embeddings is that these are not efficient across various health text applications and even less number of research describe evaluation of these embedding for health domain. In this paper, distributional embedding model is performed to acquire a word representation on data crawled from Journal of Medical Case Reports. This distributed embedding model is analyzed qualitatively and quantitatively over crawled corpus. Qualitative evaluation is employed by cosine similarity on different categories and is visually represented. Quantitative evaluation performed with parts of speech tagging and entity recognition. The embedding model attained a cross-validation accuracy of 91.70% in parts of speech tagging for GENIA corpus and ensured 83% accuracy in the entity recognition of i2b2 clinical data.

The enormous number of user activity on online social networks results in a considerable amount of data which expresses the opinion from millions of people with diversity in their social aspects. The freedom of language usage shared through social media paves the way for the existence of code-mixed data that turns out to be more complex for mining the information out of it. Considering this, we created Kannada-English code mixed corpus by crawling Facebook comments. As of now, there is no relevant corpus as well as literature available for code-mixed Kannada-English sentiment analysis. In addition to the crawled corpus, we also used sentiment analysis code-mixed corpus provided by Sentiment Analysis for Indian Languages (SAIL)-2017 which includes Bengali-English and Hindi-English languages. This paper also addresses the performance of distributed representation methods in sentiment analysis task. We have reported comparisons among different machine learning and deep learning techniques.

This paper discusses on task 1, “Affect in Tweets” sharedtask, conducted in SemEval-2018. This task comprises of various subtasks, which required participants to analyse over different emotions and sentiments based on the provided tweet data and also measure the intensity of these emotions for subsequent subtasks. Our approach in these task was to come up with a model on count based representation and use machine learning techniques for regression and classification related tasks. In this work, we use a simple bag of words technique for supervised text classification model as to compare, that even with some advance distributed representation models we can still achieve significant accuracy. Further, fine tuning on various parameters for the bag of word, representation model we acquired better scores over various other baseline models (Vinayan et al.) participated in the sharedtask.

In this paper we did an analysis of “Affects in Tweets” which was one of the task conducted by semeval 2018. Task was to build a model which is able to do regression and classification of different emotions from the given tweets data set. We developed a base model for all the subtasks using distributed representation (Doc2Vec) and applied machine learning techniques for classification and regression. Distributed representation is an unsupervised algorithm which is capable of learning fixed length feature representation from variable length texts. Machine learning techniques used for regression is ’Linear Regression’ while ’Random Forest Tree’ is used for classification purpose. Empirical results obtained for all the subtasks by our model are shown in this paper.

This paper assesses the performance of frequency and concept based text representation in Mixed Script Information Retrieval and Classification tasks. In text analytics, representation serves as an unresolved research problem to progress further towards different applications. In this paper observations from different text representation methods in text classification and information retrieval are presented. The data set from the Mixed Script Information Retrieval shared task is used in this experiment and the performance of final submitted model is evaluated by task organizers. It is observed that distributional representation performs better than the frequency based text representation methods. The final system attained first place in task 2 and was 3.89% lesser than the top scored system in task 1.

Emoji is one of the “fastest growing language ” in pop-culture, especially in social media and it is very unlikely for its usage to decrease. These are generally used to bring an extra level of meaning to the texts, posted on social media platforms. Providing such an added info, gives more insights to the plain text, arising to hidden interpretation within the text. This paper explains our analysis on Task 2, ” Multilingual Emoji Prediction” sharedtask conducted by Semeval-2018. In the task, a predicted emoji based on a piece of Twitter text are labelled under 20 different classes (most commonly used emojis) where these classes are learnt and further predicted are made for unseen Twitter text. In this work, we have experimented and analysed emojis predicted based on Twitter text, as a classification problem where the entailing emoji is considered as a label for every individual text data. We have implemented this using distributed representation of text through fastText. Also, we have made an effort to demonstrate how fastText framework can be useful in case of emoji prediction. This task is divide into two subtask, they are based on dataset presented in two different languages English and Spanish.

Emotions are a way of expressing human sentiments. In the modern era, social media is a platform where we convey our emotions. These emotions can be joy, anger, sadness and fear. Understanding the emotions from the written sentences is an interesting part in knowing about the writer. In the amount of digital language shared through social media, a considerable amount of data reflects the sentiment or emotion towards some product, person and organization. Since these texts are from users with diverse social aspects, these texts can be used to enrich the application related to the business intelligence. More than the sentiment, identification of intensity of the sentiment will enrich the performance of the end application. In this paper we experimented the intensity prediction as a text classification problem that evaluates the distributed representation text using aggregated sum and dimensionality reduction of the glove vectors of the words present in the respective texts .

Noise removal is considered to be an efficacious step in processing any kind of data. Here the proposed model deals with removal of noise from aperiodic and piecewise constant signals by utilizing wavelet transform, which is being realized in GNU Radio platform. We have also dealt with the replacement of Universal Software Radio Peripheral with RTL-SDR for a low cost Radio Frequency Receiver system without any compromise in its efficiency. Wavelet analyzes noise level separately at each wavelet scale in time-scale domain and adapts the denoising algorithm especially for aperiodic and piecewise constant signals. GNU Radio companion serves well in analysis and synthesis of real time signals.

Semantic Textual Similarity measures similarity between pair of texts, even though the similar context is projected using different words. This work attempted to incorporate the context space of the sentence from that sentence alone. It proposes combination of Word2Vec and Non-Negative Matrix Factorization to represent the sentence as context embedding vector in context space. Distance and correlation values between context embedding vector pairs used as a features for Support Vector Regression to built the domain independent similarity measuring model. The proposed model yielding performance 0.41 in terms of correlation.