Data mining

Data science projects development with Amazon SageMaker

Yuri Demchenko, Oleg Chertov, Marharyta Aleksandrova, Juan J. Cuadrado-Gallego

Big Data Infrastructure Technologies for Data Analytics

2024

This chapter discusses SageMaker - a fully managed machine learning (ML) service provided by Amazon Web Services (AWS). Being a fully managed service, means that a user does not have to deal with hardware setup, patching, management, backups etc. All this is taken care of by the service provider. The user can choose from a wide variety of computing instance types that are optimized for different tasks,

Cloud and systems

Covering a graph with dense subgraph families, via triangle-rich sets

Sabyasachi Basu, Daniel Paul Pena, Kun Qian, Seshadhri Comandur, Eddie Huang, Karthik Subbian

CIKM 2024

2024

Graphs are a fundamental data structure used to represent relation-ships in domains as diverse as the social sciences, bioinformatics, cybersecurity, the Internet, and more. One of the central observations in network science is that real-world graphs are globally sparse, yet contains numerous “pockets" of high edge density. A fundamental task in graph mining is to discover these dense sub-graphs. Most common

Information and knowledge management

A bespoke question intent taxonomy for e-commerce

Diji Yang, Omar Alonso

SIGIR 2024 Workshop on eCommerce

2024

Effective question-intent understanding plays an important role in enhancing the performance of Question-Answering (QA) and Search systems. Previous research in open-domain QA has highlighted the value of intent taxonomies in comprehending data and facilitating answer generation and evaluation. However, existing taxonomies have limitations for specific domains. We’re interested in question intent for e-commerce

Search and information retrieval

Shopping trajectory representation learning with pre-training for e-commerce customer understanding and recommendation

Yankai Chen, Tuan Truong, Xin Shen, Jin Li, Irwin King

KDD 2024

2024

Understanding customer behavior is crucial for improving service quality in large-scale E-commerce. This paper proposes C-STAR, a new framework that learns compact representations from customer shopping journeys, with good versatility to fuel multiple down-stream customer-centric tasks. We define the notion of shopping trajectory that encompasses customer interactions at the level of product categories,

Machine learning

GraphStorm: All-in-one graph machine learning framework for industry applications

Da Zheng, Xiang Song, Qi Zhu, Jian Zhang, Theodore Vasiloudis, Runjie Ma, Houyu Zhang, Zichen Wang, Soji Adeshina, Israt Nisa, Alejandro Mottini, Qingjun Cui, Huzefa Rangwala, Belinda Zeng, Christos Faloutsos, George Karypis

KDD 2024

2024

Graph machine learning (GML) is effective in many business applications. However, making GML easy to use and applicable to industry applications with massive datasets remain challenging. We developed GraphStorm, which provides an end-to-end solution for scalable graph construction, graph model training and inference. GraphStorm has the following desirable properties: (a) Easy to use: it can perform graph

Cloud and systems

A product-aware query auto-completion framework for e-commerce search via retrieval-augmented generation method

Andy Sun, Tianqi Zheng, Aakash Kolekar, Rohit Patki, Hossein Khazaei, Xuan Guo, George Cai, David Liu, Ruirui Li, Yupin Huang, Dante Everaert, Hanqing Lu, Garima Patel, Monica Cheng

SIGIR 2024 Workshop on Information Retrieval's Role in RAG Systems (IR-RAG)

2024

Query Auto-Completion (QAC) is a fundamental component of user search experience on e-commerce websites. It assists in finding userintended products, by automatically presenting search queries as users typing in the search bar. Traditional QAC systems build upon query popularity to suggest a list of potential completions, but they fall short for unforeseen search prefixes. A generative Large Language Model

Search and information retrieval

RA-NER: Retrieval augmented NER for knowledge intensive named entity recognition

Zhenwei Dai, Chen Luo, Zhen Li, Xianfeng Tang, Hanqing Lu, Rahul Goutam, Haiyang Zhang

ICLR 2024 Tiny Papers

2024

NER (named entity recognition) model aims to recognize the named entities in the keywords. However, when the entities are extremely knowledge intensive, traditional NER model cannot encode all the knowledge in its parameters, thus fails to recognize those entities with high accuracy. In this paper, we propose retrieval augmented NER model (RA-NER) to address this issue. RA-NER retrieves the most relevant

Machine learning

Bi-CAT: Improving robustness of LLM-based text rankers to conditional distribution shifts

Sriram Srinivasan, Stephen Sheng, Rishabh Deshmukh, Chen Luo, Yesh Dattatreya, Subhajit Sanyal, S. V. N. Vishwanathan

The Web Conference 2024 Workshop on Information Retrieval Meets Large Language Models

2024

Retrieval and ranking lie at the heart of several applications like search, question-answering, and recommendations. The use of Large language models (LLMs) such as BERT in these applications have shown promising results in recent times. Recent works on text-based retrievers and rankers show promising results by using bi-encoders (BE) architecture with BERT like LLMs for retrieval and a cross-attention

Search and information retrieval

NETINFOF framework: Measuring and exploiting network usable information

Meng-Chieh Lee, Haiyang Yu, Jian Zhang, Vassilis N. Ioannidis, Xiang Song, Soji Adeshina, Da Zheng, Christos Faloutsos

ICLR 2024

2024

Given a node-attributed graph, and a graph task (link prediction or node classification), can we tell if a graph neural network (GNN) will perform well? More specifically, do the graph structure and the node features carry enough usable information for the task? Our goals are (1) to develop a fast tool to measure how much information is in the graph structure and in the node features, and (2) to exploit

Machine learning

Data augmentation for supervised code translation learning

Anonymous

MSR 2024

2024

Data-driven program translation has been recently the focus of sev- eral lines of research. A common and robust strategy is supervised learning. However, there is typically a lack of parallel training data, i.e., pairs of code snippets in source and target language. While many data augmentation techniques exist in the domain of natural language processing, they cannot be easily adapted to tackle code translation

Information and knowledge management

Data mining

Work with us