Fuzzy Search

Let the search account for spelling mistakes and typos and find similar matches.

This feature module is part of the TablePress Max plan.

The “Fuzzy Search” module allows the table search to match results that are not necessarily exactly the same as the search term. Rows will be found even if a search term has typos, spelling mistakes, or is written in a dialect.

Feature description

A common example for use of fuzzy search e.g. in databases is name searching. While “Smith” and “Smythe” are pronounced in the same way, a regular search for “Smith” would not find “Smythe”, whereas a fuzzy search would.

The following examples show some possible uses.

Fuzzy search with toggle control

In this table with fictional data, a search for the name “Smith” is performed. Note how rows with similar spellings of this name are also found. By hovering the mouse over the search field, the visitor can toggle between the “exact” (on rows matching “Smith” exactly) and the “fuzzy” search (similar rows are found as well).

To test this yourself, change the search term e.g. to “Mayer”.

First Name	Last Name	ZIP	Points	Average
Gloria	Smythe	67439	4	0.1
Graham	Bonner	90727	4	0.9
Warren	Wheeler	99134	2	0.7
Zena	Hale	19803	9	0.3
Julia	Haupt	24116	10	1.0
Rachel	Smith	58951	7	0.3
Lionel	Barry	65036	7	0.5
Zena	Miller	16874	5	0.6
Dillon	Bradford	91543	7	0.4
Haley	Mcleod	99321	4	0.1
Abra	House	99537	6	0.1
Ariel	Smyth	56555	6	0.7
Roary	Ayers	12720	3	0.5
Mason	Moller	81062	3	0.6
Gillian	Simon	69137	10	0.7
Rebekah	Fisher	32572	3	0.7
Cameron	Walls	18788	2	0.2
Phelan	Meyer	95413	6	0.2
Mona	Fischer	48912	9	0.6
Rashad	Webb	27993	4	0.1
Brandon	Kinney	00986	2	0.2
Sacha	Gould	17722	8	0.7
Brenda	Osborn	52139	4	0.5
Sophia	Meier	79688	3	0.2
Bree	Orr	70160	9	0.7
Yen	Bernard	99216	1	0.3
Lacota	Le	64964	10	0.7
Nayda	Horton	42308	10	0.1
Lee	Navarro	48918	5	0.4
Margaret	Austin	26181	7	0.3
Brynne	Randolph	68523	8	0.9
Oleg	Mayer	32249	9	0.4
Mark	Cooley	54923	3	0.2
Brendan	Sharp	47511	1	0.5
Holly	Carrillo	71689	1	0.3
Kelsey	Moses	27661	5	0.8
Geraldine	Tyson	98049	3	0.6
Jasper	Maier	92329	3	0.5
Noble	Conley	14869	8	0.5
Natalie	Luna	66637	8	0.7
McKenzie	Powell	86259	9	0.4
Tyler	Dennis	13676	6	0.2
Vladimir	Jefferson	78569	7	0.3
Brett	Pena	14218	1	0.8
Demetrius	Key	67985	2	0.4
Gwendolyn	Schmitt	97324	1	0.8
Marvin	Middleton	10141	2	0.7
Nadine	Walters	27484	6	0.7
Claire	Schmidt	54218	5	0.4

Showing the similarity score

To find all rows that match when doing a “fuzzy” search, the search algorithm calculates a “similarity score”, which can be interpreted as the percentage or degree to which the search term matches the row data. Showing this to the site visitor can give hints about the confidence for the search results. This table shows an example, see the last column:

First Name	Last Name	ZIP	Points	Average
Gloria	Smythe	67439	4	0.1
Graham	Bonner	90727	4	0.9
Warren	Wheeler	99134	2	0.7
Zena	Hale	19803	9	0.3
Julia	Haupt	24116	10	1.0
Rachel	Smith	58951	7	0.3
Lionel	Barry	65036	7	0.5
Zena	Miller	16874	5	0.6
Dillon	Bradford	91543	7	0.4
Haley	Mcleod	99321	4	0.1
Abra	House	99537	6	0.1
Ariel	Smyth	56555	6	0.7
Roary	Ayers	12720	3	0.5
Mason	Moller	81062	3	0.6
Gillian	Simon	69137	10	0.7
Rebekah	Fisher	32572	3	0.7
Cameron	Walls	18788	2	0.2
Phelan	Meyer	95413	6	0.2
Mona	Fischer	48912	9	0.6
Rashad	Webb	27993	4	0.1
Brandon	Kinney	00986	2	0.2
Sacha	Gould	17722	8	0.7
Brenda	Osborn	52139	4	0.5
Sophia	Meier	79688	3	0.2
Bree	Orr	70160	9	0.7
Yen	Bernard	99216	1	0.3
Lacota	Le	64964	10	0.7
Nayda	Horton	42308	10	0.1
Lee	Navarro	48918	5	0.4
Margaret	Austin	26181	7	0.3
Brynne	Randolph	68523	8	0.9
Oleg	Mayer	32249	9	0.4
Mark	Cooley	54923	3	0.2
Brendan	Sharp	47511	1	0.5
Holly	Carrillo	71689	1	0.3
Kelsey	Moses	27661	5	0.8
Geraldine	Tyson	98049	3	0.6
Jasper	Maier	92329	3	0.5
Noble	Conley	14869	8	0.5
Natalie	Luna	66637	8	0.7
McKenzie	Powell	86259	9	0.4
Tyler	Dennis	13676	6	0.2
Vladimir	Jefferson	78569	7	0.3
Brett	Pena	14218	1	0.8
Demetrius	Key	67985	2	0.4
Gwendolyn	Schmitt	97324	1	0.8
Marvin	Middleton	10141	2	0.7
Nadine	Walters	27484	6	0.7
Claire	Schmidt	54218	5	0.4

Further details

Besides this “ranking column” that shows the similarity score and the toggle control that the site visitor can use to switch between “exact” and “fuzzy” search, the feature module also allows to configure the “threshold” at which a search results is said to be matching the search term.

All these settings can be configured in the easy-to-use TablePress user interface, individually for each table.

Usage instructions

To make this feature available for your tables, enable the “Fuzzy Search” module on the TablePress “Modules” screen.

To use this feature with your tables, activate it in the “Fuzzy Search” section on the table’s “Edit” screen. You can then configure it in the “Advanced settings” of that section.

On this page