Semi-supervised Machine Learning Approach Research Articles

Lithofacies Classification Using Supervised and Semi-Supervised Machine Learning Approach

The machine learning approach can help Geoscientists do their work in well log analysis to developing the oil and gas field. Prediction categorical or numerical response variable using a set of predictor variables supervises and semi-supervises learning is an important goal of the machine learning approach in classifying lithofacies using well log data. Semi-supervised classification offers the possibility of exploring the structure of the data without entirely external knowledge or guidance in the form of target or class information, and semi-supervised is very rarely research in the field of lithofacies classification. Well log data in gamma-ray, resistivity, neutrality, and density logs are collected and selected for data processing and transformation. The use of machine learning algorithms such as Naí¯ve Bayes, SVM, and Decision Tree is to find the log pattern or pattern classifications of lithofacies in supervised and semi-supervised to create a model with conditions requiring the change of data and the corresponding requirements. All supervised machine learning algorithms have the best accuracy because algorithms provide useful predictive in classifications based on the target but not if there are no targets given or semi-supervised. This paper compares some of the famous classification algorithms of machine learning, such as Decision tree, SVM, and Naí¯ve Bayes, on classifying lithofacies with supervised and semi-supervised learning. This research found that the semi-supervised learning of Naí¯ve Bayes has performed well in classified lithofacies. In contrast, in supervised learning, Decision Tree and SVM are superior in accuracy and visualization approach based on expert’s interpretation.

Speaking social Europe: A paradigmatic shift in the European Commission Presidents’ social policy discourse?

This article presents an original analysis of the social policy discourse at the European Union level by focusing on the language used by European Commission Presidents when addressing social issues. Through an empirical analysis based on a semi-supervised machine learning approach, we consider the entire corpus of speeches made by Barroso and Juncker from 2010 to 2018 and we classify all sentences according to three possible variants of social policy language: social-retrenchment, social-investment and rights-based language. We argue that the focus on these languages allows us to understand whether the ‘fiscal consolidation discipline’ narrative, which dominated in the aftermath of the Great Recession, has been challenged over the years by the recourse to alternative ‘social policy images’. The results confirm a significant shift in the overall configuration of social policy discourse between Barroso and Juncker. This shift has been mainly characterized by an increase in the use of rights-based language and a decrease in the use of social-retrenchment language, which however has not been completely abandoned. Therefore, the recalibration of EC’s speeches on social policy has not been abrupt but has instead been incremental. The article concludes by discussing preliminary interpretations of the observed trends and their potential implications in terms of ‘discursive legacy’ in the post-pandemic era.

A cell-to-patient machine learning transfer approach uncovers novel basal-like breast cancer prognostic markers amongst alternative splice variants

BackgroundBreast cancer is amongst the 10 first causes of death in women worldwide. Around 20% of patients are misdiagnosed leading to early metastasis, resistance to treatment and relapse. Many clinical and gene expression profiles have been successfully used to classify breast tumours into 5 major types with different prognosis and sensitivity to specific treatments. Unfortunately, these profiles have failed to subclassify breast tumours into more subtypes to improve diagnostics and survival rate. Alternative splicing is emerging as a new source of highly specific biomarkers to classify tumours in different grades. Taking advantage of extensive public transcriptomics datasets in breast cancer cell lines (CCLE) and breast cancer tumours (TCGA), we have addressed the capacity of alternative splice variants to subclassify highly aggressive breast cancers.ResultsTranscriptomics analysis of alternative splicing events between luminal, basal A and basal B breast cancer cell lines identified a unique splicing signature for a subtype of tumours, the basal B, whose classification is not in use in the clinic yet. Basal B cell lines, in contrast with luminal and basal A, are highly metastatic and express epithelial-to-mesenchymal (EMT) markers, which are hallmarks of cell invasion and resistance to drugs. By developing a semi-supervised machine learning approach, we transferred the molecular knowledge gained from these cell lines into patients to subclassify basal-like triple negative tumours into basal A- and basal B-like categories. Changes in splicing of 25 alternative exons, intimately related to EMT and cell invasion such as ENAH, CD44 and CTNND1, were sufficient to identify the basal-like patients with the worst prognosis. Moreover, patients expressing this basal B-specific splicing signature also expressed newly identified biomarkers of metastasis-initiating cells, like CD36, supporting a more invasive phenotype for this basal B-like breast cancer subtype.ConclusionsUsing a novel machine learning approach, we have identified an EMT-related splicing signature capable of subclassifying the most aggressive type of breast cancer, which are basal-like triple negative tumours. This proof-of-concept demonstrates that the biological knowledge acquired from cell lines can be transferred to patients data for further clinical investigation. More studies, particularly in 3D culture and organoids, will increase the accuracy of this transfer of knowledge, which will open new perspectives into the development of novel therapeutic strategies and the further identification of specific biomarkers for drug resistance and cancer relapse.

Image-Label Recovery on Fashion Data Using Image Similarity from Triple Siamese Network

Weakly labeled data are inevitable in various research areas in artificial intelligence (AI) where one has a modicum of knowledge about the complete dataset. One of the reasons for weakly labeled data in AI is insufficient accurately labeled data. Strict privacy control or accidental loss may also cause missing-data problems. However, supervised machine learning (ML) requires accurately labeled data in order to successfully solve a problem. Data labeling is difficult and time-consuming as it requires manual work, perfect results, and sometimes human experts to be involved (e.g., medical labeled data). In contrast, unlabeled data are inexpensive and easily available. Due to there not being enough labeled training data, researchers sometimes only obtain one or few data points per category or label. Training a supervised ML model from the small set of labeled data is a challenging task. The objective of this research is to recover missing labels from the dataset using state-of-the-art ML techniques using a semisupervised ML approach. In this work, a novel convolutional neural network-based framework is trained with a few instances of a class to perform metric learning. The dataset is then converted into a graph signal, which is recovered using a recover algorithm (RA) in graph Fourier transform. The proposed approach was evaluated on a Fashion dataset for accuracy and precision and performed significantly better than graph neural networks and other state-of-the-art methods.

Comparison of approaches for intrusion detection in substations using the IEC 60870-5-104 protocol

Electrical networks of transmission system operators are mostly built up as isolated networks without access to the Internet. With the increasing popularity of smart grids, securing the communication network has become more important to avoid cyber-attacks that could result in possible power outages. For misuse detection, signature-based approaches are already in use and special rules for a wide range of protocols have been developed. However, one big disadvantage of signature-based intrusion detection is that zero-day exploits cannot be detected. Machine-learning-based anomaly detection methods have the potential to achieve that. In this paper, various such methods for intrusion detection in substations, which use the asynchronous communication protocol International Electrotechnical Commission (IEC) 60870-5-104, are tested and compared. The evaluation of the proposed methods is performed by applying them to a data set which includes normal operation traffic and four different attacks. While the results of supervised and semi-supervised machine learning approaches are rather encouraging, the unsupervised and signature-based methods suffer from general bad performance and had difficulties to detect some attacks.

Spatial pseudo-labeling for semi-supervised facies classification

For the Quantitative classification of facies is crucial to link seismic data with its corresponding lithology for the evaluation of reservoir properties. During the past decade, seismic volumes have increased to the degree that it is perplexing for the geoscientists and seismic interpreters to examine every single seismic section to carry out facies classification. This has triggered and inspired the need and development of machine learning approaches to efficiently predict seismic facies. However, classified data or labeled data (i.e., well log data) are limited by various restraints and needs huge amount of time and cost to obtain, whereas unlabeled data (seismic data) are in abundance and relatively easy to obtain. Geoscientists are tasked with interpreting large amounts of unlabeled data using very little labeled data. In this paper, we adopt a semi-supervised machine learning approach utilizing pseudo-labeling technique to predict seismic facies and to account for the scarcity of training data by taking advantage of unlabeled data. In each iteration, a small amount of unlabeled data is labeled in spatial manner and the most confident labels (labels that are within 95% confidence interval) are added back to the training data set, starting from labeling the unlabeled data adjacent to the well and spatially moving away in small steps (few traces in each step). The labels obtained from the unlabeled data near the well (pseudo-labels) are added to the labeled data to retrain the deep learning model (classifier) in each iteration. This addition of spatial pseudo-labels adds invariance and diversity to the classifier which in turn helps classifier capitalize on known input variance, which might act as a helpful prior knowledge for next prediction. In other words, the proposed workflow accounts for spatial change in lithology that occurs while moving away from the well. Using our proposed workflow, we observed that, the classification accuracy of a machine learning model trained on a small amount of training data can be enhanced considerably. Pseudo-labels also helped improve lateral continuity of facies. Furthermore, the proposed method outperformed conventional methods, achieving a test accuracy of 96.7% and logarithmic loss as low as 0.15 compared to 91% and 0.6, respectively for supervised baseline. Whereas, for training and validation; the training accuracy, validation accuracy and loss were 98.9%, 99.09% and 0.029 respectively.

Inline Drift Detection Using Monitoring Systems and Machine Learning in Selective Laser Melting

Direct metal laser sintering, an additive manufacturing technique, has a huge growing demand in industries like aerospace, biomedical, and tooling sector due to its capability to manufacture complex parts with ease. Despite many technological advancements, the reliability and repeatability of the process are still an issue. Therefore, there is a demand for inline automatic fault detection and postprocessing tools to analyze the acquired in situ monitoring data aiming to provide better‐quality assurance to the user. Herein, the treatment of the data obtained using the EOSTATE optical tomography monitoring system is focused. A balanced dataset is obtained with the help of computer tomography of the certified part (Stainless Steel CX cylindrical samples), through which a feature matrix is prepared, and the layers of the part are classified either having “Drift” or “No‐drift.” The model is trained with the feature matrix and tested on benchmark parts (Maraging Steel) and on an industrial part (knuckle, automotive part) manufactured in AlSi10Mg. The proposed semisupervised approach shows promising results for presented case studies. Thus, the semisupervised machine learning approach, if adopted, could prove to be a cost effective and fast approach to postprocess the in situ monitoring data with much ease.

Semi-supervised Machine Learning Enables the Robust Detection of Multireference Character at Low Cost.

Multireference (MR) diagnostics are common tools for identifying strongly correlated electronic structure that makes single-reference (SR) methods (e.g., density functional theory or DFT) insufficient for accurate property prediction. However, MR diagnostics typically require computationally demanding correlated wave function theory (WFT) calculations, and diagnostics often disagree or fail to predict MR effects on properties. To overcome these challenges, we introduce a semi-supervised machine learning (ML) approach with virtual adversarial training (VAT) of an MR classifier using 15 WFT and DFT MR diagnostics as inputs. In semi-supervised learning, only the most extreme SR or MR points are labeled, and the remaining point labels are learned. The resulting VAT model outperforms the alternatives, as quantified by the distinct property distributions of SR- and MR-classified molecules. To reduce the cost of generating inputs to the VAT model, we leverage the VAT model's robustness to noisy inputs by replacing WFT MR diagnostics with regression predictions in an MR decision engine workflow that preserves excellent performance. We demonstrate the transferability of our approach to larger molecules and those with distinct chemical composition from the training set. This MR decision engine demonstrates promise as a low-cost, high-accuracy approach to the automatic detection of strong correlation for predictive high-throughput screening.

Semi-supervised regression trees with application to QSAR modelling

Despite the ease of collecting abundance of data about various phenomena, obtaining labeled data needed for learning models with high predictive performance remains a difficult and expensive task in many domains.This issue is particularly present in the case of the analysis of scientific data where obtaining labeled data typically requires expensive experiments. Moreover, in the analysis of scientific data, another issue is of fundamental importance: the interpretability of the models and the explainability of their decisions. By taking into account these considerations, we propose a novel semi-supervised method to learn regression trees. Thanks to the semi-supervised machine learning approach, the method is able to exploit information coming not only from labeled data, but also from unlabeled data, thus alleviating the issue of lack of labeled data. The method is based on the predictive clustering trees paradigm that extends regression trees towards structured output prediction. This allows us to obtain interpretable regression trees.The method we propose is particularly suited for the chemoinformatics task of quantitative structure-activity relationship (QSAR) modeling, which is the main application context considered in this paper. Specifically, we evaluate the proposed method on 4 QSAR modelling datasets and illustrate its use on a case study of predicting farnesyltransferase inhibitors. Additionally, we also evaluate our approach on 10 benchmark datasets not related to the QSAR modeling problem.The evaluation reveals the following: semi-supervised trees and ensembles thereof have better predictive performance than their supervised counterparts (especially when the number of labeled examples is very small); different datasets and different amounts of labeled data require different amounts of unlabeled data to be included in the learning process; and the learned semi-supervised regression trees can be used to better understand the problem at hand and the way predictions are being made.

Semi-Supervised Machine Learning and Adaptive Data Clustering Approach for Software Defect Prediction

Semi-Supervised Machine Learning and Adaptive Data Clustering Approach for Software Defect Prediction

A Semi-Supervised Machine Learning Approach Using K-Means Algorithm to Prevent Burst Header Packet Flooding Attack in Optical Burst Switching Network

Optical burst switching (OBS) network is a new generation optical communication technology. In an OBS network, an edge node first sends a control packet, called burst header packet (BHP) which reserves the necessary resources for the upcoming data burst (DB). Once the reservation is complete, the DB starts travelling to its destination through the reserved path. A notable attack on OBS network is BHP flooding attack where an edge node sends BHPs to reserve resources, but never actually sends the associated DB. As a result the reserved resources are wasted and when this happen in sufficiently large scale, a denial of service (DoS) may take place. In this study, we propose a semi-supervised machine learning approach using k-means algorithm, to detect malicious nodes in an OBS network. The proposed semi-supervised model was trained and validated with small amount data from a selected dataset. Experiments show that the model can classify the nodes into either behaving or not-behaving classes with 90% accuracy when trained with just 20% of data. When the nodes are classified into behaving, not-behaving and potentially not-behaving classes, the model shows 65.15% and 71.84% accuracy if trained with 20% and 30% of data respectively. Comparison with some notable works revealed that the proposed model outperforms them in many respects.

Iterative processes: a review of semi-supervised machine learning in rehabilitation science

Purpose: To define semi-supervised machine learning (SSML) and explore current and potential applications of this analytic strategy in rehabilitation research.Method: We conducted a scoping review using PubMed, GoogleScholar and Medline. Studies were included if they: (1) described a semi-supervised approach to apply machine learning algorithms during data analysis and (2) examined constructs encompassed by the International Classification of Functioning, Disability and Health (ICF). The first two authors reviewed identified articles and recorded study and participant characteristics. The ICF domain used in each study was also identified.Results: After combining information from the eight studies, we established that SSML was a feasible approach for analysis of complex data in rehabilitation research. We also determined that semi-supervised approaches may be more accurate than supervised machine learning approaches.Conclusions: A semi-supervised approach to machine learning has potential to enhance our understanding of complex data sets in rehabilitation science. SSML mirrors the iterative process of rehabilitation, making this approach ideal for calibrating devices, classifying activities or identifying just-in-time interventions. Rehabilitation scientists who are interested in conducting SSML should collaborate with data scientists to advance the application of this approach within our field.Implications for rehabilitationSemi-supervised machine learning applications may be a feasible approach for analyses of complex data sets in rehabilitation research.Semi-supervised machine learning approaches uses a combination of labelled and unlabelled data to produce accurate predictive models, thereby requiring less user-input data than other machine learning approaches (i.e., supervised, unsupervised), reducing resource cost and user-burden.Semi-supervised machine learning is an iterative process that, when applied to rehabilitation assessment and outcomes, could produce accurate personalized models for treatment.Rehabilitation researchers and data scientists should collaborate to implement semi-supervised machine learning approaches in rehabilitation research, optimizing the power of large datasets that are becoming more readily available within the field (e.g., EEG signals, sensors, smarthomes).

HMMRATAC: a Hidden Markov ModeleR for ATAC-seq.

ATAC-seq has been widely adopted to identify accessible chromatin regions across the genome. However, current data analysis still utilizes approaches initially designed for ChIP-seq or DNase-seq, without considering the transposase digested DNA fragments that contain additional nucleosome positioning information. We present the first dedicated ATAC-seq analysis tool, a semi-supervised machine learning approach named HMMRATAC. HMMRATAC splits a single ATAC-seq dataset into nucleosome-free and nucleosome-enriched signals, learns the unique chromatin structure around accessible regions, and then predicts accessible regions across the entire genome. We show that HMMRATAC outperforms the popular peak-calling algorithms on published human ATAC-seq datasets. We find that single-end sequenced or size-selected ATAC-seq datasets result in a loss of sensitivity compared to paired-end datasets without size-selection.

Personalized Maternal Sleep Quality Assessment: An Objective IoT-based Longitudinal Study

Sleep is a composite of physiological and behavioral processes that undergo substantial changes during and after pregnancy. These changes might lead to sleep disorders and adverse pregnancy outcomes. Several studies have investigated this issue; however, they were restricted to subjective measurements or short-term actigraphy methods. This is insufficient for a longitudinal maternal sleep quality evaluation. A longitudinal study: 1) requires a long-term data collection approach to acquire data from everyday routines of mothers and 2) demands a sleep quality assessment method exploiting a large volume of multivariate data to assess sleep adaptations and overall sleep quality. In this paper, we present an Internet-of-Things-based long-term monitoring system to perform an objective sleep quality assessment. We conduct longitudinal monitoring, where 20 pregnant mothers are remotely monitored for six months of pregnancy and one month postpartum. To evaluate sleep quality adaptations, we: 1) extract several sleep attributes and study their variations during the monitoring and 2) propose a semi-supervised machine learning approach to create a personalized sleep model for each subject. The model provides an abnormality score, which allows an explicit representation of the sleep quality in a clinical routine, reflecting possible sleep quality degradation with respect to her own data. Sleep data of 13 participants (out of 20) are included in our analysis, as their data are adequate for the study, including 172.15±33.29 days of sleep data per person. Our fine-grained objective measurements indicate that the sleep duration and sleep efficiency are deteriorated in pregnancy and notably in postpartum. In comparison to the mid of the second trimester, the sleep model indicates the increase of sleep abnormality at the end of pregnancy (2.87 times) and postpartum (5.62 times). We also show that the model enables individualized and effective care for sleep disturbances during pregnancy, as compared to a baseline method.

Understanding flows in high-speed scientific networks: A Netflow data study

Complex science workflows involve very large data demands and resource-intensive computations. These demands need reliable high-speed networks, that can optimize performance for application data flows. Characterizing flows into large flows (elephant) versus small flows (mice) can allow networks to optimize performance by detecting and handling demands in real-time. However, predicting elephant versus mice flows is extremely difficult as their definition varies based on networks.Machine learning techniques can help classify flows into two distinct clusters to identify characteristics of transfers. In this paper, we investigate unsupervised and semi-supervised machine learning approaches to classify flows in real time. We develop a Gaussian Mixture Model combined with an initialization algorithm, to develop a novel general-purpose method to help classification based on network sites (in terms of data transfers, flow rates and durations). Our results show that the proposed algorithm is able to cluster elephants and mice with an accuracy rate of 90%. We analyzed NetFlow reports of 1 month from 3 ESnet site routers to train the model and predict clusters.

Semi-supervised machine learning approaches for predicting the chronology of archaeological sites: A case study of temples from medieval Angkor, Cambodia.

Archaeologists often need to date and group artifact types to discern typologies, chronologies, and classifications. For over a century, statisticians have been using classification and clustering techniques to infer patterns in data that can be defined by algorithms. In the case of archaeology, linear regression algorithms are often used to chronologically date features and sites, and pattern recognition is used to develop typologies and classifications. However, archaeological data is often expensive to collect, and analyses are often limited by poor sample sizes and datasets. Here we show that recent advances in computation allow archaeologists to use machine learning based on much of the same statistical theory to address more complex problems using increased computing power and larger and incomplete datasets. This paper approaches the problem of predicting the chronology of archaeological sites through a case study of medieval temples in Angkor, Cambodia. For this study, we have a large dataset of temples with known architectural elements and artifacts; however, less than ten percent of the sample of temples have known dates, and much of the attribute data is incomplete. Our results suggest that the algorithms can predict dates for temples from 821–1150 CE with a 49-66-year average absolute error. We find that this method surpasses traditional supervised and unsupervised statistical approaches for under-specified portions of the dataset and is a promising new method for anthropological inquiry.

Predictive Model of Linear Antimicrobial Peptides Active against Gram-Negative Bacteria.

Antimicrobial peptides (AMPs) have been identified as a potential new class of anti-infectives for drug development. There are a lot of computational methods that try to predict AMPs. Most of them can only predict if a peptide will show any antimicrobial potency, but to the best of our knowledge, there are no tools which can predict antimicrobial potency against particular strains. Here we present a predictive model of linear AMPs being active against particular Gram-negative strains relying on a semi-supervised machine-learning approach with a density-based clustering algorithm. The algorithm can well distinguish peptides active against particular strains from others which may also be active but not against the considered strain. The available AMP prediction tools cannot carry out this task. The prediction tool based on the algorithm suggested herein is available on https://dbaasp.org.

APIReal: an API recognition and linking approach for online developer forums

When discussing programming issues on social platforms (e.g, Stack Overflow, Twitter), developers often mention APIs in natural language texts. Extracting API mentions from natural language texts serves as the prerequisite to effective indexing and searching for API-related information in software engineering social content. The task of extracting API mentions from natural language texts involves two steps: 1) distinguishing API mentions from other English words (i.e., API recognition), 2) disambiguating a recognized API mention to its unique fully qualified name (i.e., API linking). Software engineering social content lacks consistent API mentions and sentence writing format. As a result, API recognition and linking have to deal with the inherent ambiguity of API mentions in informal text, for example, due to the ambiguity between the API sense of a common word and the normal sense of the word (e.g., append, apply and merge), the simple name of an API can map to several APIs of the same library or of different libraries, or different writing forms of an API should be linked to the same API. In this paper, we propose a semi-supervised machine learning approach that exploits name synonyms and rich semantic context of API mentions for API recognition in informal text. Based on the results of our API recognition approach, we further propose an API linking approach leveraging a set of domain-specific heuristics, including mention-mention similarity, scope filtering, and mention-entry similarity, to determine which API in the knowledge base a recognized API actually refers to. To evaluate our API recognition approach, we use 1205 API mentions of three libraries (Pandas, Numpy, and Matplotlib) from Stack Overflow text. We also evaluate our API linking approach with 120 recognized API mentions of these three libraries.

Semi-supervised machine learning approach for DDoS detection

Even though advanced Machine Learning (ML) techniques have been adopted for DDoS detection, the attack remains a major threat of the Internet. Most of the existing ML-based DDoS detection approache...

AUTOMATIC INFORMATION EXTRACTION FROM TEXT

We present a method for automatic extract the hyponym-hypernym relations from the text data. In previous years many researchers were worked on this system but they use some pre-encoded knowledge and patterns for implementing this type of system. But this not that much use when we think about extracting more relations or discovering any new pattern. This type of system discovered one more risk which is once we use the predefined pattern and if this pattern failed to produce new pattern then all most all operation will fail due to the previous wrong pattern. The researcher was used semi-supervised machine learning approach for introducing such kind of information extraction system but this paper focuses on converting the semi-supervised machine learning approach into unsupervised machine learning approach for fully automatic extracting information from text. This paper is trying to focus on these previous issues. The paper focuses on two main objectives. (i) Avoid pre-encoded pattern for more efficiency. (ii) Define a method for automatically extracting useful relationships from an unsupervised machine learning approach. We demonstrate a machine learning approach and, especially, at different levels and in different ways, can be used to create a practical IE system. We unsupervised machine learning approach gives the better result than semi-supervised machine learning approach in term of information extraction.