mirror of
https://github.com/hanzoai/papers.git
synced 2026-10-29 07:34:31 +00:00
zeekay
5a1289e458
Leadership and team cards, board and founding prose, an AI persona modelled on a named individual, contributor and partner entries, and personal correspondence addresses all named one person. The organisation carries them now. Other people keep their names. Where a card, a board list or a founding line named someone else alongside, they stay exactly as they were; only the one entry was removed, and prose that listed two names now reads with one. Sample and fixture data uses the placeholder name these tests already use elsewhere, so the IAM username defect the notes describe still reads as written. The zoo/yapp MIT notice names the foundation that publishes it.
70 lines
2.3 KiB
TeX
70 lines
2.3 KiB
TeX
\documentclass[11pt,a4paper]{article}
|
|
\usepackage{../shared/hanzocover}
|
|
|
|
% Packages
|
|
\usepackage[utf8]{inputenc}
|
|
\usepackage[T1]{fontenc}
|
|
\usepackage{amsmath,amssymb,amsthm}
|
|
\usepackage{graphicx}
|
|
\usepackage{hyperref}
|
|
\usepackage{algorithm}
|
|
\usepackage{algpseudocode}
|
|
\usepackage{booktabs}
|
|
\usepackage{listings}
|
|
\input{../shared/lstlang}
|
|
\usepackage{xcolor}
|
|
\usepackage[numbers]{natbib}
|
|
\usepackage{tikz}
|
|
\usetikzlibrary{arrows,shapes,positioning,calc}
|
|
|
|
% Theorem environments
|
|
\newtheorem{theorem}{Theorem}[section]
|
|
\newtheorem{lemma}[theorem]{Lemma}
|
|
\newtheorem{proposition}[theorem]{Proposition}
|
|
\newtheorem{definition}[theorem]{Definition}
|
|
\newtheorem{corollary}[theorem]{Corollary}
|
|
|
|
% Code listing style
|
|
\lstset{
|
|
basicstyle=\ttfamily\small,
|
|
keywordstyle=\color{blue},
|
|
commentstyle=\color{gray},
|
|
stringstyle=\color{red},
|
|
breaklines=true,
|
|
frame=single,
|
|
numbers=left,
|
|
numberstyle=\tiny\color{gray}
|
|
}
|
|
|
|
\title{\textbf{Machine Learning for Commerce Analytics:\\Predictive Models for Customer Behavior}}
|
|
|
|
\author{
|
|
Hanzo AI Research\\
|
|
Hanzo Industries\\
|
|
\texttt{[email protected]}
|
|
}
|
|
|
|
\date{October 2018}
|
|
|
|
\begin{document}
|
|
|
|
\hanzocoverpage
|
|
|
|
\begin{abstract}
|
|
We present Hanzo Analytics, a machine learning platform for e-commerce that provides predictive analytics, customer segmentation, and churn prediction. Our system processes behavioral event streams in real-time to construct customer profiles and predict future actions. We introduce a novel feature engineering pipeline that extracts 847 features from raw event data, enabling predictive models that achieve 0.89 AUC for purchase prediction and 0.84 AUC for churn prediction. Customer segmentation using our modified k-means algorithm with behavioral embeddings improves marketing campaign ROI by 2.3x compared to demographic-only segmentation. The platform processes 50M events daily across 500+ merchants.
|
|
\end{abstract}
|
|
|
|
\input{sections/01-introduction}
|
|
\input{sections/02-system-architecture}
|
|
\input{sections/03-feature-engineering}
|
|
\input{sections/04-predictive-models}
|
|
\input{sections/05-customer-segmentation}
|
|
\input{sections/06-real-time-scoring}
|
|
\input{sections/07-model-training-pipeline}
|
|
\input{sections/08-evaluation}
|
|
\input{sections/09-privacy-and-compliance}
|
|
\input{sections/10-related-work}
|
|
\input{sections/99-conclusion}
|
|
|
|
\end{document}
|