Contradiction Detection (CD) in text is a difficult NLP task. We investigate CD over functions (e.g., BornIn(Person)=Place), and present a domain-independent algorithm that automatically discovers phrases denoting functions with high precision. Previous work on CD has investigated hand-chosen sentence pairs. In contrast, we automatically harvested from the Web pairs of sentences that appear contradictory, but were surprised to find that most pairs are in fact consistent. For example, "Mozart was born in Salzburg" does not contradict "Mozart was born in Austria" despite the functional nature of the phrase "was born in". We show that background knowledge about meronyms (e.g., Salzburg is in Austria), synonyms, functions, and more is essential for success in the CD task.

It's a contradiction - No, it's not

In recent days, the Government and other organizations are focusing on providing better health care to people. Understanding the patients experience of care-received is key for providing better health care. With prevailing usage of social media applications, patients are expressing their experience over social media. This patient authored text is a free-unstructured data which is available over social media in large chunks. To extract the sentiments from this huge data, a domain-specific dictionary is required to get better accuracy. The proposed approach defines a new domain-specific dictionary and uses this in sentiment scoring to enhance the overall sentiment classification on patient authored text. We conducted experiments on the proposed approach using NHS Choices dataset and compared it with popular classifiers like linear regression, stochastic gradient descent, dictionary-based approaches: VADER and AFINN. The results prove that the proposed approach is an effective strategy for sentiment analysis over patient authored text which helps in improving the classification accuracy.

Evolving dictionary based sentiment scoring framework for patient authored text

This study explores the effects of student behavior patterns on students’ learning performance in a student-generated question experiment. Thirty-three engineering students from a C programming course were recruited as participants. For data collection, student-generated questions, exercises, a final examination, and system logs recorded by an online student-generated questions system were analyzed. The results revealed several significant findings. Firstly, we found that the students accessed the questions generated by the classmate more frequently and for longer than other materials (i.e., exercises and lecture slides). Our results also reveal great diversity among student viewing activities; we therefore partitioned them into three behavior clusters: “Highly-engaged students” who dominate other clusters in the use of all kinds of materials, “Moderately-engaged students” who spent more time on the lecture slides and SGQ, and “Less-engaged students” who seldom used learning materials but accessed exercises more frequently than the “Moderately-engaged students” and also had the highest Test Anxiety. We also observed that motivation and learning performance are strongly associated with behavior patterns. We discuss possible explanations of our findings and propose suggestions for future research.

/pdf/effect-of-behavior-patterns-of-accessing-learning-materials-3b39k04t3s.pdf

Effect of Behavior Patterns of Accessing Learning Materials on Learning Performance in Student-generated Question Activities

Sentiment analysis is widely used to retrieve the hidden sentiments in medical discussions over Online Social Networking platforms such as Twitter, Facebook, Instagram. People often tend to convey their feelings concerning their medical problems over social media platforms. Practitioners and health care workers have started to observe these discussions to assess the impact of health-related issues among the people. This helps in providing better care to improve the quality of life. Dementia is a serious disease in western countries like the United States of America and the United Kingdom, and the respective governments are providing facilities to the affected people. There is much chatter over social media platforms concerning the patients’ care, healthy measures to be followed to avoid disease, check early indications. These chatters have to be carefully monitored to help the officials take necessary precautions for the betterment of the affected. A novel Feature engineering architecture that involves feature-split for sentiment analysis of medical chatter over online social networks with the pipeline is proposed that can be used on any Machine Learning model. The proposed model used the fuzzy membership function in refining the outputs. The machine learning model has obtained sentiment score is subjected to fuzzification and defuzzification by using the trapezoid membership function and center of sums method, respectively. Three datasets are considered for comparison of the proposed and the regular model. The proposed approach delivered better results than the normal approach and is proved to be an effective approach for sentiment analysis of medical discussions over online social networks.

Fuzzy based feature engineering architecture for sentiment analysis of medical discussion over online social networks

Across the globe, right from people who are technologically naive to people who are with high-technical knowledge, use, share, spread, and connect with each other through online social networks. People start believing and sharing the social network content without any proof of its authenticity. In several cases, the reliability of the information that gets shared between the users remains questionable due to the anonymity of the information creators. In this work, a fuzzy detection algorithm is proposed to identify the trustable content in social media. The proposed methodology is evaluated on Twitter social network platform. By computing the predictive measures, the efficiency of the proposed approach is well established.

https://ietresearch.onlinelibrary.wiley.com/doi/pdf/10.1049/iet-net.2018.5208

Madda Rajasekhara Babu

Papers

Fuzzy-based fake information detection algorithm to define the user trust on the content of social networks