000000045 001__ 45
000000045 005__ 20240801202640.0
000000045 02470 $$ahttps://doi.org/10.3886/jw31-2j38$$2DOI
000000045 037__ $$aADMIN
000000045 245__ $$aSUSTAINABLESIGNALS: An AI Approach for Inferring Consumer Product Sustainability
000000045 251__ $$av1
000000045 269__ $$a2023-05-25
000000045 336__ $$aDataset
000000045 510__ $$aLin, Tong, Xu, Tianliang, Tomkins, Sabina, and Amit, Zac. SUSTAINABLESIGNALS: An AI Approach for Inferring Consumer Product Sustainability. Inter-university Consortium for Political and Social Research [distributor], 2023-05-25. https://doi.org/10.3886/jw31-2j38
000000045 520__ $$aThe everyday consumption of household goods is a significant source of environmental pollution. As people increasingly shop online, this affords an opportunity to provide consumers with actionable feedback on the social and environmental impact of potential purchases. In our work, we explore the following questions on Amazon a) do consumers bring up the environment in their reviews either directly or through relevant related topics? b) do they tend to bring up the environment when they are satisfied or dissatisfied with a product? c) in what granular context do they bring up the environment?
<br><br>
To address these questions, we designed an annotation task and recruited knowledgeable students to annotate consumer product reviews. This dataset comprises annotations for 779 individual reviews, with each review corresponding to a distinct product.
<br><br>
In our paper, we propose a machine learning method using these annotations that can discover signals of sustainability and infer a product's sustainability score. Our model and code are released at <a href="https://github.com/Sabina321/sustainable_signals">https://github.com/Sabina321/sustainable_signals</a>. The data is for the following paper: Tong Lin, Tianliang Xu, Amit Zac, and Sabina Tomkins. SUSTAINABLESIGNALS: An AI Approach for Inferring Consumer Product Sustainability. In Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence, 2023. AI and Social Good Track.
000000045 540__ $$aThis dataset has two levels of access: Public and Login Required.
<ul>
<li>Public files can be downloaded directly from the dataset record. Typically, documentation files such as READMEs and codebooks are made public.</li>

<li>Login Required files can be downloaded directly from the dataset record, once you have created and logged in to your SOMAR account. By downloading data files you agree to <a href="https://socialmediaarchive.org/pages/?page=Terms%20of%20Use&ln=en">ICPSR’s Terms of Use</a>.</li>
000000045 650__ $$aenvironment
000000045 650__ $$ashopping
000000045 655__ $$aweb platform data
000000045 655__ $$atext
000000045 655__ $$asurvey data
000000045 720__ $$aLin, Tong$$eData Collector$$uSchool of Information, University of Michigan$$7Personal
000000045 720__ $$aXu, Tianliang$$eData Collector$$uSchool of Information, University of Michigan$$7Personal
000000045 720__ $$aTomkins, Sabina$$ePrincipal Investigator$$uSchool of Information, University of Michigan$$7Personal
000000045 720__ $$aZac, Amit$$eProject Member$$uCenter for Law and Economics, ETH Zurich$$7Personal
000000045 791__ $$tDeep learning model$$aElectronic Source$$eIs Derived From$$2URL$$whttps://github.com/Sabina321/sustainable_signals
000000045 8564_ $$yThe everyday consumption of household goods is a significant source of environmental pollution. As people increasingly shop online, this affords an opportunity to provide consumers with actionable feedback on the social and environmental impact of potential purchases. In our work, we explore the following questions on Amazon a) do consumers bring up the environment in their reviews either directly or through relevant related topics? b) do they tend to bring up the environment when they are satisfied or dissatisfied with a product? c) in what granular context do they bring up the environment?   To address these questions we designed an annotation task and recruited knowledgeable students to annotate consumer product reviews. This dataset comprises annotations for 779 individual reviews, with each review corresponding to a distinct product.   In our paper "SUSTAINABLESIGNALS: An AI Approach for Inferring Consumer Product Sustainability", we propose a machine learning method using these annotations that can discover signals of sustainability and infer a product's sustainability score. Our model and code are released at https://github.com/Sabina321/sustainable_signals. $$98ecfbea6-3dac-4248-b7eb-97e9d5e853fc$$s346752$$uhttps://socialmediaarchive.org/record/45/files/amazon%20product%20sus%20annotations.csv
000000045 8564_ $$yData summary$$908fc3207-0e29-467c-b361-e1b0ccfa1deb$$s72386$$uhttps://socialmediaarchive.org/record/45/files/Data%20Summary.pdf
000000045 907__ $$a2023-01-23$$b2023-02-20
000000045 908__ $$aAmazon.com
000000045 910__ $$aweb-based survey
000000045 911__ $$aWhen people are purchasing different categories of products on Amazon, they may pay attention to different aspects of sustainability. We are interested in the following questions: a) do consumers bring up the environment in their reviews either directly or through relevant related topics? b) do they tend to bring up the environment when they are satisfied or dissatisfied with a product? c) in what granular context do they bring up the environment? 
000000045 912__ $$aWe recruited graduate students from SEAS at the University of Michigan to annotate the customer reviews from Amazon.
000000045 923__ $$ahttps://github.com/Sabina321/sustainable_signals
000000045 924__ $$aThese annotations are used to help pre-train our deep learning models.
000000045 926__ $$aPython
000000045 927__ $$ddatasets==2.7.1,
numpy==1.20.1,
pandas==1.2.4,
scikit_learn==1.0.2,
scipy==1.6.2,
torch==1.12.1,
tqdm==4.59.0,
transformers==4.25.1
000000045 928__ $$aSynthetic data and instructions are provided in our repo.
000000045 929__ $$aYou can find the outputs of our model in the ''output'' folder, and testing results in the ''expe'' folder.
000000045 980__ $$aDatasets
000000045 980__ $$aArtificial Intelligence (AI)
000000045 981__ $$aPublished