Landemore: Response to Brennan

Despite the avalanche of facts that Jason Brennan brings against the average voter and his skepticism of deliberation among ordinary citizens, the resulting attack on democracy turns out to be surprisingly limited. In the end, Brennan concedes, democracy is still the best regime around, no matter how flawed, and we have a duty to fix it. He also concedes that deliberation among randomly selected citizens is going to be part of the solution. Landemore argues that Brennan’s solution, however—a combination of randomly selected mini-publics designing questionnaires and a weighted vote system based on these questionnaires—is still too elitist, empirically inattentive to existing experiments in deliberative democracy, and unlikely to work.

Stacked Framework for Ensemble of Heterogeneous Classification Algorithms

Ensemble methods fabricate a sequence of classifiers for classifying fresh instances by procuring a weighted vote of their individual predictions. Toning down the error and increasing accuracy is an avant-garde problem in ensemble classification. This paper presents a novel generic object-oriented voting and weighting adapted stacking framework for utilizing an ensemble of classifiers for prediction. This universal framework operates based on the weighted average of the probabilities of any suite of base learners and the final prediction is the aggregate of their respective votes. For illustrative purposes, three familiar heterogeneous classifiers, such as the Support Vector Machine, [Formula: see text]-Nearest Neighbor and Naïve Bayes, are utilized as candidates for ensemble classification using the proposed stacked framework. Further, the ensemble classifier built upon the framework is compared with others and evaluated using various cross-validation levels and percentage splits on a range of benchmark datasets. The outcome distinguishes the framework from the competition. The proposed framework is used to predict the crime propensity of prisoners most accurately, with 99.9901% accuracy.

Coarse, Efficient Decision-Making

To minimize the cost of making decisions, an agent should use criteria to sort alternatives and each criterion should sort coarsely. To decide on a movie, for example, an agent could use one criterion that orders movies by genre categories, another by director categories, and so on, with a small number of categories in each case. The agent then needs to aggregate the criterion orderings, possibly by a weighted vote, to arrive at choices. As criteria become coarser (each criterion has fewer categories) decision-making costs fall, even though an agent must then use more criteria. The most efficient option is consequently to select the binary criteria with two categories each. This result holds even when the marginal cost of using additional categories diminishes to 0. The extensive use of coarse criteria in practice may therefore be a result of optimization rather than cognitive limitations. Binary criteria also generate choice functions that maximize rational preferences: decision-making efficiency implies rational choice.

Prediksi Penyakit Jantung Menggunakan Metode-Metode Machine Learning Berbasis Ensemble – Weighted Vote

Kematian yang disebabkan penyakit jantung masih sangat tinggi, sehingga perlu peningkatan upaya-upaya pencegahannya, misalnya dengan meningkatkan capaian model prediksinya. Penerapan metode-metode machine learning pada dataset publik (Cleveland, Hungary, Switzerland, VA Long Beach, & Statlog) yang umumnya digunakan oleh para peneliti untuk prediksi penyakit jantung, termasuk pengembangan alat bantunya, masih belum menangani missing value, noisy data, unbalanced class, dan bahkan data validation secara efisien. Oleh karena itu, pendekatan imputasi mean/mode diusulkan untuk menangani missing value replacement, Min-Max Normalization untuk menangani smoothing noisy data, K-Fold Cross Validation untuk menangani data validation, dan pendekatan ensemble menggunakan metode Weighted Vote (WV) yang dapat menyatukan kinerja tiap-tiap metode machine learning untuk mengambil keputusan klasifikasi sekaligus untuk mereduksi unbalanced class. Hasil penelitian ini menunjukkan bahwa metode yang diusulkan tersebut memberikan akurasi sebesar 85,21%, sehingga mampu meningkatkan kinerja akurasi metode-metode machine learning, selisih 7,14% dengan Artificial Neural Network, 2,77% dengan Support Vector Machine, 0,34% dengan C4.5, 2,94% dengan Naïve Bayes, dan 3,95% dengan k-Nearest Neighbor.

GenRank: a R/Bioconductor package for prioritization of candidate genes

Modern high-throughput studies often yield long lists of genes, a fraction of which are of high relevance to the phenotype of interest. To prioritize the candidate genes of complex genetic traits, our R/Bioconductor package GenRank ranks genes based on convergent evidence obtained from multiple layers of independent evidence. We implemented three methods to rank genes that integrate gene-level data generated from multiple layers of evidence: (a) the convergent evidence (CE) method aggregates evidence based on a weighted vote counting method; (b) the rank product (RP) method performs a meta-analysis of microarray-based gene expression data, and (c) the traditional method combines p-values. The methods are implemented in R and are available as a package in the Bioconductor repository (http://bioconductor.org/packages/GenRank/).

Informed Traders and Balanced Books

We explore sports gambler and bookmaker behavior by examining the pregame price movements of sports contracts listed on the Tradesports betting exchange. The vast majority of prior gambling studies that examine the price efficiency of bookmaker-style betting markets and the associated informational processing ability of bettors rely on the balanced book assumption. But, on Tradesports, bet prices are directly observable and therefore we are not required to assume that the point spreads set by bookmakers are a reliable proxy for the market’s dollar-weighted vote on contest outcome. We find that pregame NBA and NFL contract price movements are reliable predictors of future contract values. The magnitude of these movements is sufficient to remove the possibility of profit-taking by tipoff/kickoff. Results are consistent with the notion that in bookmaker-style NBA betting markets there exist informed traders and that lines are set at levels that balance wagers on either side of the bet. However, NFL bookmakers may allow (or perhaps encourage) imbalanced wagering in order to exploit relatively unskilled NFL bettors.