Depression Intensity Classification from Tweets Using Fast Text Based Weighted Soft Voting Ensemble

Predicting depression intensity from microblogs and social media posts has numerous benefits and applications,including predicting early psychological disorders and stress in individuals or the general public.A major challenge in predicting depression using social media posts is that the existing studies do not focus on predicting the intensity of depression in social media texts but rather only perform the binary classification of depression and moreover noisy data makes it difficult to predict the true depression in the social media text.This study intends to begin by collecting relevant Tweets and generating a corpus of 210000 public tweets using Twitter public application programming interfaces(APIs).A strategy is devised to filter out only depression-related tweets by creating a list of relevant hashtags to reduce noise in the corpus.Furthermore,an algorithm is developed to annotate the data into three depression classes:‘Mild,’‘Moderate,’and‘Severe,’based on International Classification of Diseases-10(ICD-10)depression diagnostic criteria.Different baseline classifiers are applied to the annotated dataset to get a preliminary idea of classification performance on the corpus.Further FastText-based model is applied and fine-tuned with different preprocessing techniques and hyperparameter tuning to produce the tuned model,which significantly increases the depression classification performance to an 84%F1 score and 90%accuracy compared to baselines.Finally,a FastText-based weighted soft voting ensemble(WSVE)is proposed to boost the model’s performance by combining several other classifiers and assigning weights to individual models according to their individual performances.The proposed WSVE outperformed all baselines as well as FastText alone,with an F1 of 89%,5%higher than FastText alone,and an accuracy of 93%,3%higher than FastText alone.The proposed model better captures the contextual features of the relatively small sample class and aids in the detection of early depression intensity prediction from tweets with impactful performances.

Hyperchaos and MD5 Based Efficient Color Image Cipher

While designing and developing encryption algorithms for text and images,the main focus has remained on security.This has led to insufficient attention on the improvement of encryption efficiency,enhancement of hyperchaotic sequence randomness,and dynamic DNA-based S-box.In this regard,a new symmetric block cipher scheme has been proposed.It uses dynamic DNA-based S-box connected with MD5 and a hyperchaotic system to produce confusion and diffusion for encrypting color images.Our proposed scheme supports various size color images.It generates three DNA based Sboxes for substitution namely DNA_1_s-box,DNA_2_s-box and DNA_3_sbox,each of size 16×16.Next,the 4D hyperchaotic system followed by MD5 is employed in a novel way to enhance security.The three DNAbased S-boxes are generated from real DNA sequences taken from National Center for Biotechnology Information(NCBI)databases and are dependent on the mean intensity value of an input image,thus effectively introducing content-based confusion.Finally,Conservative Site-Specific Recombination(CSSR)is applied on the output DNA received from DNA based S-boxes.The experimental results indicate that the proposed encryption scheme is more secure,robust,and computationally efficient than some of the recently published similar works.Being computational efficient,our proposed scheme is feasible on many emergent resource-constrained platforms.

Decision Level Fusion Using Hybrid Classifier for Mental Disease Classification

Mental health signifies the emotional,social,and psychological well-being of a person.It also affects the way of thinking,feeling,and situation handling of a person.Stable mental health helps in working with full potential in all stages of life from childhood to adulthood therefore it is of significant importance to find out the onset of the mental disease in order to maintain balance in life.Mental health problems are rising globally and constituting a burden on healthcare systems.Early diagnosis can help the professionals in the treatment that may lead to complications if they remain untreated.The machine learning models are highly prevalent for medical data analysis,disease diagnosis,and psychiatric nosology.This research addresses the challenge of detecting six major psychological disorders,namely,Anxiety,Bipolar Disorder,Conversion Disorder,Depression,Mental Retardation and Schizophrenia.These challenges are mined by applying decision level fusion of supervised machine learning algorithms.A dataset was collected from a clinical psychologist consisting of 1771 observations that we used for training and testing the models.Furthermore,to reduce the impact of a conflicting decision,a voting scheme Shrewd Probing Prediction Model(SPPM)is introduced to get output from ensemble model of Random Forest and Gradient Boosting Machine(RF+GBM).This research provides an intuitive solution for mental disorder analysis among different target class labels or groups.A framework is proposed for determining the mental health problem of patients using observations of medical experts.The framework consists of an ensemble model based on RF and GBM with a novel SPPM technique.This proposed decision level fusion approach by combining RF+GBM with SPPM-MIN significantly improves the performance in terms of Accuracy,Precision,Recall,and F1-score with 71\%,73\%,71\%and 71\%respectively.This framework seems suitable in the case of huge and more diverse multiclass datasets.Furthermore,three vector spaces based on TF-IDF(unigram,bi-gram,and tri-gram)are also tested on the machine learning models and the proposed model.

A New Attempt to Neutrosophic Soft Bi-Topological Spaces

In this article,new generalized neutrosophic soft∗b open set is introduced in neutrosophic soft bi-topological structurers(NSBTS)concerning soft points of the space.This new set is produced by making the marriage of soft semi-open set with soft pre-open set in neutrosophic soft topological structure.An ample of results are investigated in NSBTS on the basis of this new neutrosophic soft∗b open set.Proper examples are settled for justification of these results.The non-validity of some results is vindicated with examples.

A New Approach to Vague Soft Bi-Topological Spaces

Fuzzy soft topology considers only membership value.It has nothing to do with the non-membership value.So an extension was needed in this direction.Vague soft topology addresses bothmembership and non-membership values simultaneously.Sometimes vague soft topology(single structure)is unable to address some complex structures.So an extension to vague soft bi-topology(double structure)was needed in this direction.To make this situation more meaningful,a new concept of vague soft bi-topological space is introduced and its structural characteristics are attempted with a new definition.In this article,new concept of vague soft bi-topological space(VSBTS)is initiated and its structural behaviors are attempted.This approach is based on generalized vague soft open sets,known as vague softβopen sets.An ample of examples are given to understand the structures.For the non-validity of some results,counter examples are provided to pay the price.Pair-wise vague softβopen and pair-wise vague softβclose sets are alsoaddressedwith examples in(VSBTS).Vague soft separation axioms are initiatedin(VSBTS)concerning soft points of the space.Other separation axioms are also addressed relative to soft points of the space.Vague soft bi-topological properties are studied with the application of vague softβopen sets with respect to soft points of the spaces.The characterization of vague softβclose as well as vague softβopen sets,characteristics of Bolzano Weirstrass property,vague soft compactness and its marriage with sequences,interconnection between sequential compactness and countable compactness in(VSBTS)with respect to softβopen sets are addressed.

Chaos-Based Cryptographic Mechanism for Smart Healthcare IoT Systems

Smart and interconnected devices can generate meaningful patient data and exchange it automatically without any human intervention in order to realize the Internet of Things(IoT)in healthcare(HIoT).Due to more and more online security and data hijacking attacks,the confidentiality,integrity and availability of data are considered serious issues in HIoT applications.In this regard,lightweight block ciphers(LBCs)are promising in resourceconstrained environment where security is the primary consideration.The prevalent challenge while designing an LBC for the HIoT environment is how to ascertain platform performance,cost,and security.Most of the existing LBCs primarily focus on text data or grayscale images.The main focus of this paper is about securing color images in a cost-effective way.We emphasis high confidentiality of color images captured by cameras in resource-constrained smartphones,and high confidentiality of sensitive images transmitted by low-power sensors in IoT systems.In order to reduce computational complexity and simulation time,the proposed Lightweight Symmetric Block Cipher(LSBC)exploits chaos-based confusion-diffusion operations at the inter-block level using a single round.The strength of LSBC is assessed by cryptanalysis,while it is ranked by comparing it to other privacy-preserving schemes.Our results show that the proposed cipher produces promising results in terms of key sensitivity and differential attacks,which proves that our LSBC is a good candidate for image security in HIoT.