EXPRESS: Bayesian Consumer Profiling: How to Estimate Consumer Characteristics from Aggregate Data

Firms use aggregate data from data brokers (e.g., Acxiom, Experian) and external data sources (e.g., Census) to infer the likely characteristics of consumers in a target list and thus better predict consumers’ profiles and needs unobtrusively. We demonstrate that the simple count method most commonly used in this effort relies implicitly on an assumption of conditional independence that fails to hold in many settings of managerial interest. We develop a Bayesian profiling introducing different conditional independence assumptions. We also show how to introduce additional observed covariates into this model. We use simulations to show that in managerially relevant settings, the Bayesian method will outperform the simple count method, often by an order of magnitude. We then compare different conditional independence assumptions in two case studies. The first example estimates customers’ age on the basis of their first names; prediction errors decrease substantially. In the second example, we infer the income, occupation, and education of online visitors of a marketing analytic software company based exclusively on their IP addresses. The face validity of the predictions improves dramatically and reveals an interesting (and more complex) endogenous list-selection mechanism than the one suggested by the simple count method.

Evaluation of Accuracy and Precision of the Sound-Recorder-Based Point-Counts Applied in Forests and Open Areas in Two Locations Situated in a Temperate and Tropical Regions

The point-count method is one of the most popular techniques for surveying birds. However, the accuracy and precision of this method may vary across various environments and geographical regions. We conducted sound-recorder-based point-counts to examine the accuracy and precision of the method for bird biodiversity estimation as a function of geographical region, habitat type and the time of day at which the survey began. In temperate (Poland) and tropical (Cameroon) regions, we recorded soundscapes on two successive mornings at 36 recording sites (18 in each location). At each site, we analyzed three 5-min surveys per day. We found no differences in the accuracy and precision of the method between regions and habitats. The accuracy was significantly greater at sunrise than during later surveys. The similarity of the bird assemblages detected by different surveys did not differ between regions or habitats. However, the bird communities described at the same time of day were significantly more similar to each other than those detected by surveys conducted at different times. The point-count method provided statistically indistinguishable estimates of bird biodiversity in different geographical regions and habitats. However, our results highlight two weaknesses of the method: low accuracy (41–54%), which limits the usefulness of a single survey in understanding bird–environment relationships, and changes in accuracy throughout the day, which may result in the misinterpretation of the status of bird populations.

A Variable Ranking Method for Machine Learning Models with Correlated Features: In-Silico Validation and Application for Diabetes Prediction

When building a predictive model for predicting a clinical outcome using machine learning techniques, the model developers are often interested in ranking the features according to their predictive ability. A commonly used approach to obtain a robust variable ranking is to apply recursive feature elimination (RFE) on multiple resamplings of the training set and then to aggregate the ranking results using the Borda count method. However, the presence of highly correlated features in the training set can deteriorate the ranking performance. In this work, we propose a variant of the method based on RFE and Borda count that takes into account the correlation between variables during the ranking procedure in order to improve the ranking performance in the presence of highly correlated features. The proposed algorithm is tested on simulated datasets in which the true variable importance is known and compared to the standard RFE-Borda count method. According to the root mean square error between the estimated rank and the true (i.e., simulated) feature importance, the proposed algorithm overcomes the standard RFE-Borda count method. Finally, the proposed algorithm is applied to a case study related to the development of a predictive model of type 2 diabetes onset.

Age-Specific Differences in Asian Elephant Defecation, Dung Decay, Detection and Their Implication for Dung Count

The various components of the dung count method of population estimation were evaluated such as defecation rate, decay rate, detection probability of dung and age-specific estimates of elephant density based on dung size. The defecation rate of elephants was determined in captive elephants of the Mudumalai elephant camp, Tamil Nadu, India. A total of 14 elephants in the dry season (Dec-Mar 2002) and 17 elephants in the wet season (Jun-Oct 2007) of different age-sex classes were observed for 42 days and 51 days by focal sampling methods and circumference of largest dung pile were measured to determine the growth curve. Total and age-specific elephant density based on dung circumference were estimated using indirect dung count method, 24 transect lines of 2 ~ 4km resulting in a total of 125km distance in Mudumalai Tiger Reserve. The dung decay rate was determined by marking fresh dung piles (n = 1628) every month with an average of 125 ± 77 dung piles/month from Jan 2007 to Feb 2008. The mean defecation rate was 13.51±0.51 (n = 94) per day. The defecation rate varied across age-sex classes and seasons with a lower defecation rate for younger age classes. The dung circumference measurements showed similar growth curves to other Asian and African elephants. Dung size-based age classification significantly overestimates the adult age class, giving an estimate of 82% adults in the population against the true 48.5%. Experimental analysis of detection of dung across perpendicular distance detection of larger and smaller dung size/age classes is equal at a visible distance and declined significantly for younger age/size class at the furthest distance. Decay rate varied significantly according to age classes and across seasons, with lower dung survival rates of younger age classes compared to adults. The minimum required sample size for the dung decay experiment was 250. The age-specific estimate of elephant density based on the indirect dung count method revealed a precise estimate on density, both decay and defecation contribute less than 12% of the variance of the estimate. Age composition based on dung count is highly skewed towards adults, younger age class were low due to lower defecation and faster decay and lower probability of detection in the transect.

Angiographic flow velocity predicts lower limb outcomes after endovascular therapy: Application of the frame count method

Introduction: Several factors related to lesion characteristics and endovascular therapy (EVT) procedures have been reported to affect primary patency after EVT. However, it is unknown why these factors were associated with primary patency. We hypothesized patency failure was related to poor blood flow in affected arteries. Methods: This retrospective study included 131 consecutive patients who had received EVT with bare metal stents for peripheral artery disease caused by femoropopliteal artery lesions. Based on the tertile post-EVT flow velocity of the superficial femoral artery (SFA), patients were divided into high ( n = 43), middle ( n = 44), and low ( n = 44) flow velocity groups. Flow velocity was measured using the frame count method. We measured incidence of major adverse limb events (MALE), composed of target lesion revascularization (TLR), non-TLR, and major amputation. Results: At a median follow-up period of 22.7 months, MALE had occurred in 7 (16.3%), 10 (22.7%), and 29 (65.9%) of patients from the high, middle, and low SFA flow velocity groups, respectively ( p < 0.001). Kaplan–Meier analysis showed incidence of MALE was significantly higher in the patients of low SFA flow velocity (log-rank test χ2 = 38.8, p < 0.001). Multivariate analysis found low SFA flow velocity to be an independent predictor for MALE (hazard ratio: 4.42; 95% CI: 2.27 to 8.60; p < 0.001) as was ankle–brachial index. Conclusion: Post-EVT SFA flow velocity for femoropopliteal artery lesions treated with bare metal stents is an independent predictor of limb patency. The frame count method for assessing arterial flow velocity is convenient and has potential for wide applications in EVT.

Analysis Of The Effect Of Age On Medication Adherence With Oral Antidiabetes And Oral Antihypertension Drugs Using Pill Count Method

Compliance with taking medication is a problematic patients who get polypharmacy prescribing, this is because of the many drugs that must be consumed. As a result of the consumption of many drugs that cause sufferers of degenerative diseases are not compliant to take medication and health awareness decreases. This is not only experienced in elderly and geriatric patients but in adulthood also experience non-compliance with taking medication due to the many activities, lazy to take medication and also forget. This study was conducted to analyze the effect of age on adherence to taking oral antidiabetic drugs and oral antihypertensive drugs using the pill count method at the Regional Hospital of South Surabaya. The Pill Count method was chosen in this study with the aim of analyzing patient compliance in taking medication by calculating the remaining drug based on the rules of use and dosage when the patient returns to control. In addition, this study also aims to determine the effect of age on patient compliance in taking oral antidiabetic drugs and oral antihypertensive drugs. This research is descriptive observational with prospective data collection. Data analysis used descriptive and continued with statistical analysis using Chi-Square. The results of this study that adherence to take medication for elderly patients who received oral antidiabetic drug therapy obtained as much as 96.7% and 90% in adult patients. Compliance to oral antihypertension drugs in adult patients are 90% and 83.3% in elderly patients. Age does not effect compliance patients with Chi-Square analysis with p value = 0,301 and 0,448 > 0.05.

Assessing The Habitat Suitability of Six Coniferous Forests Using Avian Assemblages At High Altitude Ecosystem

Birds are habitat specialist, i.e. often selects habitat that offers ample services such as a variety of food resources, safe breeding and less disturb sites to satisfy their needs. We aimed to ascertain the conservation status of avian assemblages inhabited in six different coniferous forests. The study employed the point count method to determine the avian assemblages in each forest type. The surveys were conducted employing the distance sampling point count method and determining the floristic composition through direct visual estimation from January 2018 to March 2019. In total, 2465 bird individuals representing 57 species and 29 families were observed in six coniferous forests. Twelve tree species, nine shrub species, and 14 grass species were identified in the forests. Except for the Streptopelia turtur (vulnerable), all bird species are listed as “least concern” according to the IUCN Red List Category and Criteria 2019–2. Notably, the relative abundance of bird assemblages among six coniferous habitats was significantly different (F5, 293 = 22.8, p < 0.05). The CAP results reflected that bird diversity indices varied coniferous forests. Strikingly, dry–temperate coniferous forest (BG) was densely populated (5.831 ± 0.694 birds•ha−1) as compared to other habitats. These findings suggested that subtropical coniferous forests are cradles of life for a wide array of avian assemblages and hence should be declared as protected forests to enhance avian assemblages in this region.