Showing 1-8 of 25 results
Multimodal Benchmark Dataset

Multimodal Benchmark Dataset

weimiao0112
Open Dataset
Creators: Wang, X. (Shane), Bendle, N., & Pan, Y. (2024)
Publication Date: Published by the authors in 2024
Creators: Wang, X. (Shane), Bendle, N., & Pan, Y. (2024)

To benchmark the machine learning capabilities of AWS, Google Cloud Platform, and Microsoft Azure for marketing tasks, the authors ran several publicly available unstructured datasets (image, video, and audio) through each platform. The supporting data and code are publicly available on GitHub. The data comprise three datasets: an image dataset, a video dataset, and an audio dataset.

The image dataset consists of three labelled datasets used to evaluate image classification and object detection performance: (1) a vehicle dataset containing 17,760 images labelled as vehicle or no vehicle; (2) an apparel dataset containing 11,385 images with 24 labels formed by crossing six colours (black, blue, brown, green, red, and white) with four clothing types (dress, pants, shirt, and shoes); and (3) a face authenticity dataset containing 2,041 images labelled as real or fake. The image datasets were curated from third-party datasets available on Kaggle and assembled and re-shared for this study by Wang, Bendle, and Pan.

The video dataset consists of three publicly available datasets used to evaluate video content classification on a custom model: (1) the Fight Scenes dataset, containing fighting and non-fighting actions; and two subsets of a public video advertisement dataset labelled according to persuasive strategies, including (2) exciting versus not exciting and (3) funny versus not funny (excluding private or unavailable videos). The Fight Scenes dataset is publicly available, while the advertising video subsets were derived from Hussain et al. (2017), Automatic Understanding of Image and Video Advertisements (IEEE CVPR). These datasets were assembled for this study by Wang, Bendle, and Pan.

The audio dataset consists of 3,021 podcast summaries used to evaluate speech-to-text transcription performance. Each platform generates a transcript together with confidence scores and alternative word predictions. The audio data were collected online and released by the authors through their GitHub repository.

Text-Based Measures of Firm Position and Differentiation in Technology Space

Text-Based Measures of Firm Position and Differentiation in Technology Space

weimiao0112
Open Dataset
Creators: Sam Arts; Bruno Cassiman; Jianan Hou (built on the DISCERN database by Ashish Arora, Sharon Belenzon & Lia Sheer)
Publication Date: 2023
Creators: Sam Arts; Bruno Cassiman; Jianan Hou (built on the DISCERN database by Ashish Arora, Sharon Belenzon & Lia Sheer)

Data and code that represent each firm–year technology portfolio as a vector built from the stemmed technical keywords in the firm’s patents (title, abstract, claims). Cosine similarity between vectors gives a text-based measure of how similar two firms are technologically, and of how differentiated a firm is. Built on DISCERN, which links roughly 1.35 million U.S. patents (1980–2015) to Compustat firms.

GitHub – Missing Data Handling Code for Machine Learning Portfolios

GitHub – Missing Data Handling Code for Machine Learning Portfolios

Lisa
Open Dataset
Creators: Andrew Y. Chen, Jack McCoy
Publication Date: 2024
Creators: Andrew Y. Chen, Jack McCoy

This GitHub repository contains the code implementation for the study “Missing Values Handling for Machine Learning Portfolios” (Chen & McCoy, 2024, Journal of Financial Economics). It provides replication code for the various missing data handling methods evaluated in the paper, enabling researchers to apply and compare these techniques in their own machine learning portfolio research.

Replication Data: Missing Values Handling for Machine Learning Portfolios

Replication Data: Missing Values Handling for Machine Learning Portfolios

Lisa
Open Dataset
Creators: Andrew Chen , Jack McCoy
Publication Date: 20 February 2024
Creators: Andrew Chen , Jack McCoy

This dataset, hosted on Mendeley Data, contains the replication data for the study “Missing Values Handling for Machine Learning Portfolios” (Chen & McCoy, 2024, Journal of Financial Economics). It includes stock return and firm characteristic data used to examine how different methods for handling missing values in financial datasets affect the performance of machine learning-based portfolio construction strategies, providing guidance on best practices for missing data imputation in empirical asset pricing research.

R Package for Self-Organizing Map-Based Data Imputation

R Package for Self-Organizing Map-Based Data Imputation

Lisa
Open Dataset
Creators: CRAN / missSOM package authors
Publication Date: May 5, 2022
Creators: CRAN / missSOM package authors

The missSOM package for R provides functions for imputing missing data using Self-Organizing Maps (SOM), a neural network-based approach for pattern recognition and data completion.

Sign In

New to BERD@NFDI? Register

Forgot password?

Register

Reset Password

Please enter your username or email address, you will receive a link to create a new password via email.