A large amount of materials science knowledge is generated and stored as text published in peer-reviewed scientific literature. While recent developments in natural language processing, such as Bidirectional Encoder Representations from Transformers (BERT) models, provide promising information extraction tools, these models may yield suboptimal results when applied on materials domain since they are not trained in materials science specific notations and jargons. Here, we present a materials-aware language model, namely, MatSciBERT, trained on a large corpus of peer-reviewed materials science publications. We show that MatSciBERT outperforms SciBERT, a language model trained on science corpus, and establish state-of-the-art results on three downstream tasks, named entity recognition, relation classification, and abstract classification. We make the pre-trained weights of MatSciBERT publicly accessible for accelerated materials discovery and information extraction from materials science texts.

No abstract

The three most basic amenities required for the survival of a human being are food, shelter and clothing. In today’s tech-savvy generation, the latter two have witnessed a huge scientific boost. Unfortunately, even today, agriculture is considered as more of a man-power oriented field. Most of the farmers are untutored and have little to no scientific knowledge of farming. So, they have to rely on the hit and trial method to learn from experience which leads to wastage of time and resources. Our system focuses on building a predictive model to recommend the most suitable crops to grow in a particular farm based on various parameters. This can be helpful for the farmers to be more productive and competent without wasting any resources by farming the most competent crops.

Augmenting pre-trained language models with knowledge graphs (KGs) has achieved success on various commonsense reasoning tasks. Although some works have attempted to explain the behavior of such KG-augmented models by indicating which KG inputs are salient (i.e., important for the model's prediction), it is not always clear how these explanations should be used to make the model better. In this paper, we explore whether KG explanations can be used as supervision for teaching these KG-augmented models how to filter out unhelpful KG information. To this end, we propose SALKG, a simple framework for learning from KG explanations of both coarse (Is the KG salient?) and fine (Which parts of the KG are salient?) granularity. Given the explanations generated from a task's training set, SALKG trains KG-augmented models to solve the task by focusing on KG information highlighted by the explanations as salient. Across two popular commonsense QA benchmarks and three KG-augmented models, we find that SALKG's training process can consistently improve model performance. 1 * Equal contribution. † Work done while TG interned remotely at USC. 1 Code will be released at github.com/INK-USC/SalKG.

We are facing a perennial pandemic, so as to achieve a viable livelihood number of merchants started preferring online shopping over traditional markets with significant advancements. Online markets provide everyone a seamless experience of shopping from comfort at your own home which aligns with the current pandemic scenario. We have analysed the concept of online shopping through the eyes of the consumer and carried out a market research on the present e-commerce trends. This research paper allows us to perform an analysis over a broad demography of people and deduce their preferences on E-commerce giants. We will be analysing the trends being followed in the online market and the outlook of people on diverse circumstances. The study indicates the relevance of consumer’s gratification with the features and facilities provided by various e-commerce giants.