In this paper, we apply and evaluate several machine learning and deep learning methods, along with various feature extraction and word-embedding techniques, on a consolidated dataset of 20600 instances, for hate speech detection from tweets and comments in Hinglish. The experimental results reveal that deep learning models perform better than machine learning models in general. Among the deep learning models, the CNN-BiLSTM model with word2vec word embedding provides the best results. The model yields 0.876 accuracy, 0.830 precision, 0.840 recall and 0.835 F1-score. These results surpass the recent state-of-art approaches.

Iron, the fourth most prevalent element in the earth's crust, is found in low quantities in the soil and in dissolved form to a limited extent in groundwater and the ocean. It plays an important role in oxygen and electron transport system, is present in various enzymes like catalase, oxidase, reductases, dehydrogenases and dehydrases42. Various oxidation states were shown out of which Fe (II) and Fe (III)28 are the common oxidation states. Iron plays a crucial role as its deficiency leads to anemia whereas higher amount leads to increase in risk for liver damage, cancer, cardiovascular disease, arthritis, diabetes etc. Human being contains about 4 grams of iron in the form of hemoglobin in the blood where it is needed for tissue respiration. Various methods have been reported in literature by which iron can be estimated in micro amounts like atomic absorption spectroscopy (AAS), titrimetric, voltammetry, flame atomic absorption spectrometry, chromatography, flow injection analysis, gravimetry and photometry. They have certain limitations in application. Spectrophotometric methods are also available which have an edge over the existing methods in terms of accuracy, sensitivity, selectivity and application, yet there is need of discovery of simpler spectrophotometric methods.

In this paper, we apply and evaluate several machine learning and deep learning methods, along with various feature extraction and word-embedding techniques, on a consolidated dataset of 20600 instances, for hate speech detection from tweets and comments in Hinglish. The experimental results reveal that deep learning models perform better than machine learning models in general. Among the deep learning models, the CNN-BiLSTM model with word2vec word embedding provides the best results. The model yields 0.876 accuracy, 0.830 precision, 0.840 recall and 0.835 F1-score. These results surpass the recent state-of-art approaches.

Introduction: Endometriosis is an oestrogen dependent gynaecological disease, having endometrial glands and stromal tissues outside the intrauterine locations. The etiology and pathogenesis of endometriosis is still unclear and it affects a large proportion of reproductive age women. It’s a heterogeneous disease and is found to be associated with hormonal and histological alterations. Studies indicate that mutation in aromatase (CYP19) gene is involved in a number of inflammatory diseases and CYP19 rs 2470152 site polymorphism may help to find its relation to susceptibility to endometriosis. Aim: To ascertain the relationship between changes in histological architecture in endometrial cells during endometriosis with circulating hormone levels, stress parameters and aromatase (CYP19A1) gene polymorphism {Single Nucleotide Polymorphism (SNP) rs 2470152}. Materials and Methods: This was a hospital-based casecontrol study where all patients and controls were recruited from the Outpatient Department (OPD) of the Sir Sunderlal Hospital, Department of Obstetrics and Gynecology, Institute of Medical Sciences, Banaras Hindu University (BHU), Varanasi, India from March 2016-March 2019, and a total 300 subjects, 120 endometriosis patients and 180 healthy controls were studied. Histological studies were done by Haematoxyline-Eosin (H&E) staining in the endometrial tissues of patients and controls. Genotyping of SNP rs 2470152 was conducted by Polymerise Chain Reaction-Restriction Fragment Length Polymorphism (PCR-RFLP) method on genomic Deoxyribonucleic Acid (DNA) isolated from patients and control blood. Student’s t-test was used to compare the mean for the two independent groups. Allele and genotype distribution among groups were evaluated using the Chi-squared test and Fisher’s exact test. Results: In the present study, all the subjects were in the age group of 20-50+ years where 20-40 years age group were premenopausal and 40-50+ year were perimenopausal. Significant histological changes were observed in the endometrial glands and stroma of the endometrium tissues of the diseased women compared to the healthy controls. Various pathological entities were altered in circulating blood plasma of patients than to control. For polymorphism studies allele (T,C) (p=0.002*) and genotypic (TT, TC and TT) (p<0.001*) frequencies were found significantly variable in endometriosis patients in comparison of controls. Conclusion: This study showed that endometrial tissue undergoes a lot of pathological changes during a disease and this may be due to significantly altered expression of aromatase gene leading to higher oestrogen level, causing this disease and its proliferation. Aromatase (CYP19A1) gene polymorphism was found significantly associated, and other factors may be affecting aromatase directly or indirectly in steroidogenic pathway