Utilizing data sampling techniques on algorithmic fairness for customer churn prediction with data imbalance problems.

Background: Customer churn prediction (CCP) refers to detecting which customers are likely to cancel the services provided by a service provider, for example, internet services. The class imbalance problem (CIP) in machine learning occurs when there is a huge difference in the samples of the positive class compared to the negative class. It is one of the major obstacles in CCP as it deteriorates performance in the classification process. Utilizing data sampling techniques (DSTs) helps to resolve the CIP to some extent. Methods: In this paper, we review the effect of using DSTs on algorithmic fairness, i.e., to investigate whether the results pose any discrimination between male and female groups and compare the results before and after using DSTs. Three real-world datasets with unequal balancing rates were prepared and four ubiquitous DSTs were applied to them. Six popular classification techniques were utilized in the classification process. Both classifier's performance and algorithmic fairness are evaluated with notable metrics. Results: The results indicated that the Random Forest classifier outperforms other classifiers in all three datasets and, that using SMOTE and ADASYN techniques causes more discrimination in the female group. The rate of unintentional discrimination seems to be higher in the original data of extremely unbalanced datasets under the following classifiers: Logistics Regression, LightGBM, and XGBoost. Conclusions: Algorithmic fairness has become a broadly studied area in recent years, yet there is very little systematic study on the effect of using DSTs on algorithmic fairness. This study presents important findings to further the use of algorithmic fairness in CCP research.

Assessment of biofertilizer use for sustainable agriculture in the Great Mekong Region.

A growing concern on the deleterious effects of chemical inputs to the environment has been on the rise from the excessive use of chemical inputs leading to soil and water pollution, destruction to fauna and microbial communities, reduced soil fertility and increased crop disease susceptibility. In the Great Mekong Region (GMR), a large majority of the population relies on agriculture and faces severe challenges including decline in soil fertility, increased pests and diseases, leading to lower ecosystem productivity. In this region, over-dependence on chemical fertilizers also continues to impact negatively on soil health and the wider ecosystem. Agroecological practices, and beneficial microorganisms in particular, offer an affordable and sustainable alternative to mineral inputs for improved plant nutrition and soil health for optimal crop performance and sustainable production. Biofertilizers are a key component in integrated nutrient management as well as for increased economic benefits from reduced expenditure on chemical fertilizers, holistically leading to sustainable agriculture. To cope with the need for biofertilizer adoption for sustainable agricultural production, the countries in the GMR are putting efforts in promoting development and use of biofertilizers and making them available to farmers at affordable costs. Despite these efforts, farmers continue to use chemical fertilizers at high rates with the hope of increased yields instead of taking advantage of microbial products capable of providing plant nutrients while restoring or improving soil health. This study explored the current agricultural practices in the six countries in the GMR (China, Vietnam, Myanmar, Thailand, Cambodia and Lao PDR), the critical need for sustainable agroecological practices with a special emphasis on biofertilizers. We highlighted the current status, distribution, adoption and gaps of biofertilizer production in the GMR, in order to obtain an insight on the nature of biofertilizers, efficacy and production standards, adoption or lack of biofertilizers in the GMR.

Spent coffee ground as renewable energy source: Evaluation of the drying processes.

Spent coffee ground (SCG) is an environmental nuisance material, but, if appropriately processed it can be converted into pellets, and thus, used as an energy source. The moisture content of the final product should be below 10%, to ensure safe storage, and elimination of microorganism growth (particularly moulds). The present study aims to identify the optimal drying process for removing moisture from SCG and to investigate changes to the composition of SCG due to drying, at temperatures around 75 °C, so that the dried SCG to qualify as renewable energy source. Three drying processes were employed for SCG drying (with initial moisture content of about 65%): oven drying, solar drying and open air sun drying, while SCG samples were placed in aluminium trays with thicknesses of 1.25, 2.5 and 4 cm. Based on the experimental results for SCG samples with thickness 2.5 cm, the open air sun drying process required 10 h to reach final moisture content of 37%, while solar drying achieved 10% moisture content in 10 h and oven drying achieved 7% moisture content in 6 h. The solar drying process proved as the most advantageous, due to low energy requirements and adequate quality of dried SCG. Also, experiments indicated that SCG storage at "normal room conditions" resulted to equilibrium moisture content in SCG of 8%, regardless of the initial moisture content. Furthermore, instrumental analyses of the SCG, revealed changes to its composition for a number of chemical groups, such as aldehydes, ketones, phytosterols, alkaloids, lactones, alcohols, phenols, pyrans and furans, among others. It was also identified that the SCG colour was affected due to the drying process.

FGFR4 polymorphic variants modulate phenotypic features of Cushing disease.

Cushing disease is a potentially lethal condition resulting from hormone excess, usually due to a small pituitary tumor that fails to respond to negative feedback inhibition. A minority of patients develop larger, more aggressive tumors of the same lineage but with modest hormone excess. Here we show that a common polymorphism in the fibroblast growth factor receptor 4 (FGFR4) transmembrane domain yields receptor isoforms with distinct properties that mediate these biological differences. Forced expression of the major FGFR4-G388 variant allele supports pY-signal transducer and activator of transcription (STAT3) responses. In contrast, expression of the minor FGFR4-R388 allele enhances STAT3 serine phosphorylation, driving cellular growth. In addition, FGFR4-R388 enhances glucocorticoid receptor phosphorylation and nuclear translocation. Consistent with these findings, glucocorticoid administration resulted in enhanced hormone negative feedback in mice with knock-in of the FGFR4 variant allele. Moreover, clinical data from patients with pituitary tumors revealed that those homozygous for the R388 allele have a higher frequency of silent corticotroph macroadenomas than FGFR4-G388 carriers, who were more likely to have small but hormonally active microadenomas. These findings demonstrate that the FGFR4 transmembrane polymorphic variants can modulate cellular growth and sensitivity to glucocorticoid hormone negative feedback through distinct STAT3 modifications of relevance to the human forms of Cushing disease.

The cancer-associated FGFR4-G388R polymorphism enhances pancreatic insulin secretion and modifies the risk of diabetes.

The fibroblast growth factor receptor 4 (FGFR4)-R388 single-nucleotide polymorphism has been associated with cancer risk and prognosis. Here we show that the FGFR4-R388 allele yields a receptor variant that preferentially promotes STAT3/5 signaling. This STAT activation transcriptionally induces Grb14 in pancreatic endocrine cells to promote insulin secretion. Knockin mice with the FGFR4 variant allele develop pancreatic islets that secrete more insulin, a feature that is reversed through Grb14 deletion and enhanced with FGF19 administration. We also show in humans that the FGFR4-R388 allele enhances islet function and may protect against type 2 diabetes. These data support a common genetic link underlying cancer and hyperinsulinemia.

Migrants' access to antiretroviral therapy in Thailand.

OBJECTIVE: To investigate migrants' access to antiretroviral therapy (ART) and assess the applicability of ART guidelines to migrants. METHODS: Six focus group discussions (FGD) were conducted in Thailand with 74 Burmese migrants: factory workers in Mae Sot and Bangkok, construction site workers in Chiang Mai and unemployed and undocumented HIV-positive migrants in Mae Sot. Thirteen key stakeholders and migrants were interviewed for triangulation. RESULTS: (1) Present criteria for in-/exclusion restrict migrants' access to ART. (2) Leading ART guidelines are not applicable for migrants in general. (3) Migrants are likely to experience more problems with adherence to ART than local patients, which increases the importance of ART guidelines. CONCLUSIONS: Without ART guidelines that take into consideration the specific circumstances that limit migrants' access to ART, health care providers will continue to render HIV-positive migrants ineligible. Interventions are needed to both make the ART guidelines applicable to migrants and to overcome obstacles restricting migrants' access to ART. This will greatly improve migrants' access to ART and help to save the lives of thousands of HIV-positive migrants.