Automated Mapping of Product Features Mined From Online Customer Reviews to Engineering Product Characteristics

Abstract

Until now, translating product features expressed in the market into quantifiable engineering metrics has primarily been a manual process. This manual process establishes product features from large-scale customer feedback using a product’s components from large-scale design specifications. This process exacerbates the complexity and sheer amount of information that designers must handle during the early stages of new product development. The methodology proposed in this paper automatically identifies product features by mapping terms that describe product features from technical descriptions and customer reviews. In order to discover terms related to the features expressed in the market, the authors of this work employ WordNet and the PageRank algorithm, which search for semantically similar terms in products’ technical descriptions. A case study demonstrates the methodology’s viability for matching product features that are extracted from online customer reviews to the technical descriptions that address them.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 7 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.