TypeSafe Jev vs RapidFuzz vs Sentence Transformers for Duplicate Records

A CodeCut benchmark test found that TypeSafe Jev, a fixed-answer classification model, can catch duplicate product records that RapidFuzz and Sentence Transformers miss, such as "Haier 13' Silver Tube TV - HTR13" and "13in Crt Tv W/atsc Tuner - HTR13," which share model number HTR13 but no wording. The test used the Abt-Buy entity-resolution dataset of 1,081 Abt products, 1,092 Buy.com products and 1,097 known matches, with rapidfuzz v3.14.6, sentence-transformers v5.6.0, typesafe-sdk v0.7.5 and pandas v2.3.3. The author positions Jev as cheaper than running an LLM on every product because it picks from fixed answers rather than generating text, while still allowing custom matching rules.

Table of Contents Introduction In a previous article https://codecut.ai/text-similarity-fuzzy-matching-guide/ , I compared four text similarity tools, from regex to Sentence Transformers. Fuzzy matching handled typos and word order, and embeddings handled different words with the same meaning: Jon Smith John Smith <- typo, fuzzy matching catches it NYC New York City <- abbreviation, embeddings catch it One case none of them handled well is two records that describe the same product in different words: Haier 13' Silver Tube TV - HTR13 13in Crt Tv W/atsc Tuner - HTR13 <- same model HTR13 , different words An LLM could decide what makes two products the same, for example with PydanticAI forcing a yes or no answer https://codecut.ai/enforce-structured-outputs-from-llms-with-pydanticai/ , but running one on every product would be slow and expensive. TypeSafe Jev https://docs.typesafe.ai/models is a cheaper option: it picks from fixed answers instead of writing a reply, and you can still give it your own rules. This article tests whether Jev can catch the mistakes RapidFuzz and Sentence Transformers make, and what it costs to run. 💻 Get the Code: The full script and setup are in the GitHub companion folder https://github.com/khuyentran1401/codecut-blog/tree/main/match duplicate records . Stay Current with CodeCut Easy-to-digest articles on Python, AI, and open-source tools. Delivered twice a week. .codecut-subscribe-wrap { display: flex; justify-content: center; } .codecut-subscribe-btn { background: 72BEFA important; color: 2F2D2E important; border: none; border-radius: 8px; padding: 12px 28px; font-family: inherit; font-size: 16px; font-weight: 700; cursor: pointer; text-decoration: none important; display: inline-flex; align-items: center; justify-content: center; transition: background 0.3s ease; } .codecut-subscribe-btn:hover, .codecut-subscribe-btn:focus { background: 5aa8e8 important; color: 2F2D2E important; text-decoration: none important; } / Mobile responsive / @media max-width: 480px { .codecut-subscribe-btn { width: 100%; text-align: center; } } Setup Install the libraries: pip install rapidfuzz sentence-transformers typesafe-sdk pandas python-dotenv This article uses rapidfuzz v3.14.6, sentence-transformers v5.6.0, typesafe-sdk v0.7.5, and pandas v2.3.3. The Abt-Buy benchmark https://dbs.uni-leipzig.de/research/projects/benchmark-datasets-for-entity-resolution has two product lists, 1,081 products from Abt the first list and 1,092 from Buy.com the second list , plus a file of the 1,097 known matches. I use it to count each tool’s correct and wrong merges. first list 1,081 second list 1,092 Samsung ... SBH500 ────── Samsung SBH500 ... Sony ... CDPCX455 ────── Sony CDP-CX455 ... ... Sony CDP-CX355 ... ... The code below shows only the call that matters for each tool. The data download and the full code are in the companion folder https://github.com/khuyentran1401/codecut-blog/tree/main/match duplicate records . Match Names That Look Alike with RapidFuzz RapidFuzz https://github.com/rapidfuzz/RapidFuzz is a fast fuzzy string matching library. It scores how similar two strings look from 0 to 100. John Smith vs Smith, John 100 same words John Smith vs Jon Smith 95 one letter off John Smith vs Mary Jones 50 different names Let’s start with an easy case for RapidFuzz: the same product, with the same words in a different order. First list: Samsung Stereo Bluetooth Headset In Black - SBH500 Right match: Samsung SBH500 Stereo Bluetooth Headset Ask RapidFuzz for the closest match in the second list: from rapidfuzz import fuzz, process, utils def closest match product name: str - tuple: best name, score, = process.extractOne product name, second list.name, scorer=fuzz.token set ratio, processor=utils.default process, lowercase and strip punctuation return best name, score best name, score = closest match "Samsung Stereo Bluetooth Headset In Black - SBH500" print f"Closest match: {best name} score {score:.1f} " OutputClosest match: Samsung SBH500 Stereo Bluetooth Headset score 100.0 Every word in the second list’s name also appears in the first list’s name, so RapidFuzz gives it a perfect score and finds the right match. Now try a harder product, where the right match adds many extra words: First list: GE GSD4000NWW White Built-In Dishwasher - GSD4000WH Right match: GE GSD4000NWW Dishwasher with 5 Wash Cycles STD TUB ESTAR White - gsd4000nww They are the same product because both carry the model number GSD4000NWW. product name = "GE GSD4000NWW White Built-In Dishwasher - GSD4000WH" right match = "GE GSD4000NWW Dishwasher with 5 Wash Cycles STD TUB ESTAR White - gsd4000nww" best name, best score = closest match product name right score = fuzz.token set ratio product name, right match, processor=utils.default process print f"Closest match: {best name} score {best score:.1f} " print f"Right match: {right match} score {right score:.1f} " OutputClosest match: LG Dishwasher score 87.0 Right match: GE GSD4000NWW Dishwasher with 5 Wash Cycles STD TUB ESTAR White - gsd4000nww score 75.9 The right match shares more words, but it also adds more extra words, so it scores lower. Merging every product in the first list with its closest match gives these results: Merged Correct Wrong 1,081 740 341 Nearly 1 in 3 merges is wrong. Match Meaning with Sentence Transformers Sentence Transformers https://github.com/huggingface/sentence-transformers turns text into a list of numbers, called an embedding, so texts with similar meaning end up with similar numbers. The similarity between two embeddings runs from 0 to 1. numbers = similarity to "cell phone" ★ cell phone ● mobile phone 0.87 ● laptop 0.44 ● banana 0.31 Because Sentence Transformers compares meaning instead of letters, it might do better on the dishwasher example that RapidFuzz got wrong: First list: GE GSD4000NWW White Built-In Dishwasher - GSD4000WH Right match: GE GSD4000NWW Dishwasher with 5 Wash Cycles STD TUB ESTAR White - gsd4000nww Ask Sentence Transformers for the closest match: from sentence transformers import SentenceTransformer, util model = SentenceTransformer "all-MiniLM-L6-v2" second list vectors = model.encode utils.default process name for name in second list.name query = model.encode utils.default process "GE GSD4000NWW White Built-In Dishwasher - GSD4000WH" best = util.semantic search query, second list vectors, top k=1 0 0 print f"Closest match: {second list.name best 'corpus id' } similarity {best 'score' :.2f} " OutputClosest match: GE GSD4000NWW Dishwasher with 5 Wash Cycles STD TUB ESTAR White - gsd4000nww similarity 0.83 As expected, Sentence Transformers finds the right match. This is because it rewards shared details such as the brand, model number, and color. Here is how much each detail adds: similarity to: GE GSD4000NWW White Built-In Dishwasher Dishwasher 0.65 + GE 0.75 +0.10 brand matches + GSD4000NWW 0.91 +0.16 model number matches + White 0.95 +0.04 color matches Now let’s try a product where a wrong match shares more words than the right one. I expect Sentence Transformers to pick the wrong one: First list: Sony 400 Disc MegaStorage CD Changer - CDPCX455 Right match: Sony CDP-CX455 CD Player/Changer - CDPCX455 <- same model number Wrong match: Sony CDP-CX355 300 Disc MegaStorage CD Changer <- more words in common Run Sentence Transformers and show its top 2 matches: query = model.encode utils.default process "Sony 400 Disc MegaStorage CD Changer - CDPCX455" for hit in util.semantic search query, second list vectors, top k=2 0 : print f"{hit 'score' :.2f} {second list.name hit 'corpus id' }" Output0.93 Sony CDP-CX355 300 Disc MegaStorage CD Changer 0.86 Sony CDP-CX455 CD Player/Changer - CDPCX455 As I expected, Sentence Transformers picks the wrong match. Merging every product in the first list with its closest match gives these results: 1,081 763 318 Still nearly 1 in 3 merges is wrong, only slightly better than RapidFuzz. Ask Jev Which Product Is the Same TypeSafe Jev https://docs.typesafe.ai/models is a decision model. Instead of generating text, it answers a question with fixed answers and returns the probability of each one. For example, here is the same question asked to a chat model and to Jev: review: "Cheaper than the store, though the left earbud crackles." question: What is this review about? product quality, price, shipping Chat model Claude Haiku 5.5 : Product quality. The review's complaint is that the left earbud crackles. ... Jev: product quality 0.79 <- picked price 0.21 shipping 0.00 Unlike RapidFuzz and Sentence Transformers, Jev: Follows a rule you give it, so you can specify what “the same product” means in your data Answers with a probability, so you can match the pairs with high confidence and review the rest Runs fast and cheap, so checking thousands of pairs is practical Start with the CD changer that Sentence Transformers got wrong: First list: Sony 400 Disc MegaStorage CD Changer - CDPCX455 Right match: Sony CDP-CX455 CD Player/Changer - CDPCX455 Let’s give Jev the product and 3 similar names from the second list, plus “none of these”. We will also specify what makes two products the same: the same model number, capacity, and color. from typesafe sdk import Choice, TypeSafeClient client = TypeSafeClient model="jev-latest" def ask jev product name: str, shortlist: list - dict: response = client.system one state=product name, questions={"match": Choice instructions= "Which candidate is the same product? Same model, same capacity or size, " "and same color or variant. Different model numbers mean different products." , criteria={name: None for name in shortlist + "none of these" }, }, return response.choices "match" .probabilities cd changer = ask jev "Sony 400 Disc MegaStorage CD Changer - CDPCX455", "Sony CDP-CX355 300 Disc MegaStorage CD Changer", "Sony CDP-CX455 CD Player/Changer - CDPCX455", "Sony CDP-CE375 CD Player/Changer - CDPCE375", , for name, probability in cd changer.items : print f"{probability:.2f} {name}" Output0.00 Sony CDP-CX355 300 Disc MegaStorage CD Changer 0.93 Sony CDP-CX455 CD Player/Changer - CDPCX455 0.00 Sony CDP-CE375 CD Player/Changer - CDPCE375 0.07 none of these Jev gives the right match the highest probability, 0.93. This means it follows the rule we gave it: different model numbers mean different products. Let’s compare this result with the scores from the other two tools: Candidate RapidFuzz Sentence Transformers Jev Sony CDP-CX455 CD Player/Changer right match 73.8 0.86 0.93 Sony CDP-CX355 300 Disc MegaStorage CD Changer 94.5 0.93 0.00 Sony CDP-CE375 CD Player/Changer 62.8 0.83 0.00 Only Jev ranks the right match first. It is also confident the other two are wrong. What does Jev do when none of the choices is the right match? Let’s find out by giving it three wrong candidates: First list: Sony Stereo Audio Receiver - STRDE197 Candidates: Sony STR-DG720 A/V Receiver - STRDG720 <- not a match Sony STR-DG920 A/V Receiver - STRDG920 <- not a match Sony STR-DG820 A/V Receiver - STRDG820 <- not a match receiver = ask jev "Sony Stereo Audio Receiver - STRDE197", "Sony STR-DG720 A/V Receiver - STRDG720", "Sony STR-DG920 A/V Receiver - STRDG920", "Sony STR-DG820 A/V Receiver - STRDG820", , for name, probability in receiver.items : print f"{probability:.2f} {name}" Output0.00 Sony STR-DG720 A/V Receiver - STRDG720 0.00 Sony STR-DG920 A/V Receiver - STRDG920 0.00 Sony STR-DG820 A/V Receiver - STRDG820 1.00 none of these Jev picks “none of these”, which is the right answer. RapidFuzz and Sentence Transformers always return a closest name, so they would merge a wrong one. Combine Sentence Transformers and Jev We can’t give Jev all 1,092 names from the second list as answers, because a Choice question takes at most 255. So let’s use Sentence Transformers to keep the 3 closest names for each product, and Jev chooses among them. 1,092 names in the second list │ Sentence Transformers ▼ 3 closest names │ Jev ▼ 1 match, or none Because Jev returns a probability, each product can be merged, left unmatched, or sent to a person to review. Top answer "none of these"? ── yes ──▶ leave unmatched │ no ▼ Probability above 0.7? ── yes ──▶ merge │ no ▼ Probability below 0.3? ── yes ──▶ leave unmatched │ no ▼ send to a person Running this on every product in the first list gives these results: Route Pairs Correct Wrong Merge p 0.7 808 807 1 Send to a person 62 Leave unmatched 211 Jev merges 808 pairs with only one wrong. The run took about 35 seconds and cost about $0.03 for 1,081 questions. With one wrong merge out of 808, the merges are reliable enough to apply automatically. Which Approach Should You Use? Here are the three approaches side by side: Approach Merged Correct Wrong merges RapidFuzz 1,081 740 341 Sentence Transformers 1,081 763 318 Top 3 + Jev p 0.7 808 807 1 RapidFuzz and Sentence Transformers are free, but about 1 in 3 of their merges is wrong, so you still have to check every one of them. Jev costs about $0.03 for the whole list and makes only one wrong merge, so you only need to look at the few products it is unsure about. This means you end up saving a lot by spending a little. So should you always use Jev? No. I recommend starting with the free tools to see how well they work for your data, and moving to more complicated tools only if the current ones are not good enough. As a rule of thumb: Use RapidFuzz when names look almost the same but differ in typos, case, punctuation, or word order. Use Sentence Transformers when names use different words for the same product, and no near-identical products are in the list. Use Jev when you need rules to decide if two products are the same, such as matching the model number, capacity, or color. Related Tutorials Structured Output Tools for LLMs Compared https://codecut.ai/structured-llm-outputs-tools-comparison/ : Get fixed, typed answers from a chat model Auto-Optimize an LLM Classifier with DSPy https://codecut.ai/dspy-auto-optimize-llm-classifier/ : Build a classifier from prompts and let DSPy tune them langextract vs spaCy https://codecut.ai/langextract-vs-spacy-entity-extraction-comparison/ : AI-powered against rule-based extraction of entities from text The post TypeSafe Jev vs RapidFuzz vs Sentence Transformers for Duplicate Records https://codecut.ai/typesafe-jev-vs-rapidfuzz-sentence-transformers/ appeared first on CodeCut https://codecut.ai .