Welcome to Match Studio
Welcome to Match Studio, an interactive tool for evaluating and configuring Babel Street Match for record matching. Match Studio uses Babel Street Match for fuzzy retrieval and matching, while storing the records and search keys in the Elasticsearch full-text search engine.
Match Studio includes the following options:
Search: Perform searches or batch searches, returning matches from an index. Configure search parameters, import search data, and switch between multiple indices.
Compare: Displays the details of a pairwise match, including the algorithms used to calculate the match scores. Modify the values of match parameters and see the impact on the match score. Use these values to optimize match parameters for your data and use case.
Evaluate: Calculate the accuracy of Babel Street Match using your gold data and determine the best match threshold.
Configure: In this section, you can create, save, edit, import, and export parameter configurations. A parameter configuration is a saved collection of values for the parameters that control how a match is scored. You can also use this section to manage stop words and overrides.
Server: Represented by the status icon. Click Configure Servers to access the Configure Servers page, where you can add and remove external servers or change which server Match Studio is connected to.
Help: Represented by the question mark icon. Displays this help file and version information.
Your business determines your specific use case and priorities. Search can be optimized for your use case by managing the trade-offs between accuracy and speed, as well as precision (percentage of returned results that are relevant) and recall (percentage of relevant results returned). Optimizing for recall can increase false positives; optimizing for precision can increase false negatives (missed matches).
System requirements will depend on the size of your index, the required throughput, and your target accuracy levels.
Guided tours
Match Studio includes a set of guided tours that walk you through the features and functions of Match Studio. These tours are an easy way to learn how to use the product.
To start a guided tour, click on the lightbulb on the lower left-hand corner of the product and select a tour. There are tours for Search, Compare, and Configure.
You must be on the default server to access the guided tours.
The guided tours are not available in locked mode.
Match Studio Limited (free trial)
The free trial is a limited edition of Match Studio that lets you try the tool with your own data for 5 days.
Feature | Full Product | Limited (free trial) |
|---|---|---|
Languages | Expanded language support (see Language support for full list) | English, Arabic, Chinese, Japanese Korean, Russian |
Index support | Unrestricted | Up to 3 indices at a time (including permanent OFAC list index) with up to 10,000 records per index |
Supported file types for index creation | .csv, .tsv, .xml, .json | .csv, .tsv |
Searches per day | Unlimited | 500 searches per day. Each row of a batch search file contributes to this limit. |
Parallel batch jobs | ✓ | X |
Compare addresses | ✓ | X |
Importing, editing, and deleting parameter configurations | ✓ | X |
Adding, editing, and deleting stop words | ✓ | X |
Adding, editing, and deleting overrides | ✓ | X |
Adding new servers | ✓ | X |
Unlocked mode and locked mode
At various points throughout this guide, you may see references to unlocked mode and locked mode. When you install Match Studio, you may do so in unlocked mode or locked mode. The differences between the mode are as follows:
Unlocked mode: Enables full access to Match's configurations (including parameters, stop words, and overrides) at the cost of some performance. This mode is useful for testing or in non-production environments.
Locked mode: Limits configurability from within Match Studio (configuration access is still available through Match directly). This ensures production environments cannot be modified through Match Studio, and boosts performance.
For more information on installing Match Studio in either mode, see the installation guide included with your product delivery.
Overview of matching
Matching refers to the process of comparing identifying information about an individual, such as their name, company, address, and/or age, between two records. With Match Studio, you can enter one or more pieces of identifying information and Match Studio will return a list of potential matches from your loaded index. Each match will have a score, between 0% and 100%, indicating the match strength.
Name matching is the core of multi-field entity matching. Names are complex to match because of the large number of variations that occur within a language and across languages. These include, but are not limited to, typographical errors, phonetic spelling variations, transliteration differences, initials, and nicknames.
Match Studio also matches other data types such as organization name, location name, date and address.
You can investigate why particular fields matched and how scores were calculated using the compare functionality. You can even change how the scores are calculated by modifying the match parameters in real time to better understand the process and tune it for your specific application.
Language support
Match Studio currently has two levels of language support: complete and limited. Complete support uses the full set of algorithms to calculate match scores and match parameters. The table below lists the languages and scripts with complete support. Match Studio also supports cross language support, matching between different languages.
Important
Match Studio Limited (free trial) only supports English, Arabic, Chinese, Japanese, Korean, and Russian.
For all other languages, Match Studio has limited support:
Exact matches return a score of 1. This is the same for all languages.
A score is calculated based on string edit distance.
Types of token and name matches
Match type | Example |
|---|---|
Phonetic similarity | Haylee ↔ Hailey ↔ Hayleigh |
Initials | J.E. Smith ↔ James Earl Smith |
Transliteration spelling differences | Abdul Rasheed ↔ Abd al-Rashid |
Nicknames | Wiliam ↔ Will ↔ Bill ↔ Billy |
Missing spaces or hyphens | MaryEllen ↔ Mary Ellen ↔ Mary-Ellen |
Titles and honorifics | Dr. ↔ Mr. ↔ Ph.D |
Truncated name components | McDonalds ↔ McDonald ↔ McD |
Missing name components | Phillip Charles Carr ↔ Phillip Carr |
Out-of-order name components | Diaz, Carlos Alfonzo ↔ Carlos Alfonzo Diaz |
Names split inconsistently across database fields | Dick. Van Dyke ↔ Dick Van . Dyke |
Same name in multiple languages | Mao Zedong ↔ Мао Цзэдун ↔ 毛泽东 ↔ 毛澤東 |
Semantically similar names | Eagle Pharmaceuticals, Inc. ↔ Eagle Drugs, Co. |
Semantically similar names across languages | Nippon Telegraph and Telephone Corporation ↔ 日本電信電話株式会社 |