Fastest
Eureka on the web
When you want the answer in the next five minutes.
The agent works the prompt against patents and technical literature, citing every source.
Run your analysis now →https://eureka.patsnap.com/resources/blog/rd-blog/bayesian-network-structure-learning-from-data-patent-landscape/ · Patsnap · data cut-off 2026-10-15 · downloaded from the live page
Patent Landscape · Artificial Intelligence & Machine LearningBayesian Network Structure Learning from Data: Patent Landscape
A 116-family dataset on Bayesian network structure learning: who files, the IPC split and how filings trend over time.
163
Published Records
116
Simple Patent Families
5%
Families Naming the Largest Filer
38%
In Force at Cut-off
163 published records collapse into 116 simple (DOCDB) families; the difference is the same invention filed in more than one country. INTEL CORP is named on 6 of the 116 patent families in scope. One patent family can name several filers, so these figures are not slices of a divided whole and do not add up. In force = 62 of 163 records still granted and maintained at the data cut-off; the rest are pending, withdrawn, refused or lapsed.
- 116 simple patent families and 163 published records sit in scope for this search, with no date limit applied to the query itself.
- Intel Corp holds the largest single filer position, named on 6 of the 116 families (5.2%).
- 84 of the 90 listed filers hold 2 or fewer families each, a count that shows how spread out the filer list is rather than concentrated at the top.
Published byPatsnap Research··7 min readSourced from Patsnap Eureka
Overview
What this dataset covers
This landscape is built from a search for "bayesian network*" combined with "structure learning" or "structure discovery" and a data or graph term, across patent titles, abstracts and claims. The result set contains 116 simple patent families (de-duplicated across countries) and 163 published records (one per publication), with no filing-date limit set on the query. Coverage spans every matching record the search string returns, and the filing-trend chart shows a recent window rather than the full span.
Of the 163 published records, 62 (38%) were granted and still in force at the point this data was retrieved. The remainder are pending, withdrawn, refused, lapsed or not separately determinable from this view, so that breakdown is not shown here.
Records vs. families in scope
Counted over all 163 published records, including filings from before 2007; the filing-trend chart below covers only the 144 records from 2007 onward, so the two charts are not the same set of documents. Ordered by volume only, not by importance or claim strength. Counted in simple (DOCDB) patent families, so one invention filed in several countries counts once. Names come from Patsnap's standardised assignee field as recorded on each document; subsidiaries and group companies are not merged. The icon before each name is read from the name itself (words such as University, Institute or Ltd); the data carries no organisation-type field, so treat it as a reading aid rather than a classification. Filings held by named individuals rather than organisations are removed where they can be identified (10 here); that identification is not exhaustive, so an individually held filing can still appear below. Either way those filings stay inside the totals above. This table shows the top 30 of the 90 filers returned for this search; the data endpoint returns at most 100 entities, so even that 90 is not the complete set of filers. A single family can name more than one filer, so these counts overlap and do not add up to the total.
Source: Patsnap Eureka. Assignee ranking and totals. Derived from one Patsnap search for Bayesian Network Structure Learning from Data Patent Landscape, covering all matching records regardless of filing date (the search string sets no date limit), retrieved 2026-10-15. Counts cover published records only and move as new filings publish.Run this in Eureka MCP
Let an AI agent run this analysis on your own technology
Pick a task. Every answer cites the patents behind it.
10,000 free credits to start
Data
Filing trend and technology composition
The two views below use different units: the filing trend counts published records by earliest filing year, while the technology split counts simple patent families against IPC subclass.
Filing trend by earliest filing year (published records) (144 of 163 published records)
Published records by earliest filing year run from 2 in 2007 up to a peak of 22 in 2021, with 6 recorded so far for 2026. Filings from 2022 onward are still filling in as later-filed applications continue to publish, so the most recent years understate their eventual totals; this is a count of documents, not inventions, since one invention can publish more than once across jurisdictions.
Counted in published records by earliest filing year, not de-duplicated into families. A filing is counted in the year it was filed, but it only becomes visible after it publishes: roughly 18 months for most routes, and up to 30 months where a PCT application enters national phase. 2022 onwards are therefore still filling in and are shaded; no growth figure on this page uses them. Because this series counts documents rather than inventions, a rise can mean more inventions or the same inventions filed in more countries; this dataset cannot separate the two. This chart covers 2007-2026 (144 records); the 163 total quoted above includes earlier filings outside the window.
Technology composition by IPC subclass (simple patent families)
G06N (computing based on AI models) appears on 65.5% of the 116 families, the highest share of any subclass, followed by G06F (electric digital data processing) at 50.9% and G06Q (business, commerce & admin data processing) at 25.0%. Because a family can carry more than one IPC class, these shares add up to more than 100%, and 27 subclasses appear across the full dataset.
Shares are the percentage of the 116 patent families in scope. One patent family can carry several IPC classes, so the shares add up to more than 100%.
Source: Patsnap Eureka. Filing trend and technology composition. Derived from one Patsnap search for Bayesian Network Structure Learning from Data Patent Landscape, covering all matching records regardless of filing date (the search string sets no date limit), retrieved 2026-10-15. Counts cover published records only and move as new filings publish.
Go deeper on Bayesian Network Structure Learning from Data Patent Landscape with Eureka
This page is one run against one query. Ask Eureka your own question about bayesian network structure learning from data patent landscape and every answer comes back with the patent numbers behind it.
Try EurekaPatent Example
A record from the result set
Example record
Reinforcement learning method and apparatus (US20230037632A1)
HUAWEI TECHNOLOGIES CO., LTD.
The abstract describes a reinforcement learning method and recognition apparatus: obtaining a structure graph containing structure information about an environment or an intelligent agent, learned previously; feeding the current state of the environment and that structure graph into a policy function implemented as a graph neural network; using the policy function to generate an action in response to the current state and the structure graph; outputting that action to the environment through the intelligent agent; and obtaining the next state from the environment in response.This is the first record returned by the database's default ordering, not a ranking by citations, grant status or breadth. It is shown to illustrate one way structure-graph concepts appear in filings in this set, not as representative of the field.
Selected as the first record the search returns under the database default order, then shown in full. It is not ranked by citations, claim breadth or commercial importance, and it is not a judgement about which filing in this space matters most.
Source: Patsnap Eureka. The first record the search returned, shown in full. Derived from one Patsnap search for Bayesian Network Structure Learning from Data Patent Landscape, covering all matching records regardless of filing date (the search string sets no date limit), retrieved 2026-10-15. Counts cover published records only and move as new filings publish.Run this in Eureka MCP
Run it yourself
Put your own technology through the same analysis
Where to run it
For builders
MCP server & REST API
When it has to run inside your own pipeline.
Patent search, landscape analysis and assignee resolution as MCP tools. Drop them into any agent framework, or call REST directly.
Browse MCP servers →Insights
Reading the filer and technology data
These cards report figures drawn directly from the dataset. They describe what was counted, not what it means for any one filer's strategy or standing.
Largest filer
6 families / 5.2%
of 116 simple patent families
Intel Corp named on 6 of 116 families
Intel Corp is named on 6 of the 116 simple patent families in scope, a 5.2% share. This is the single-row figure the page shows; filer rows are not summed elsewhere on this page because one family can name more than one filer.
Largest single-row share in the filer list
Filer spread
84 of 90
listed filers
84 of 90 listed filers hold 2 or fewer families
Of the 90 organisation-only filers returned by this search, 84 hold 2 or fewer simple patent families each. The remaining listed filers, including the leader at 6 families, account for the rest of the filer list; the ranking table on the page shows only the top 30 of the 90.
Count of filers at the low-volume end of the list
Technology split
65.5% / 50.9% / 25.0%
of 116 simple patent families
G06N, G06F and G06Q lead the IPC composition
G06N (computing based on AI models) appears on 65.5% of the 116 families, G06F (electric digital data processing) on 50.9%, and G06Q (business, commerce & admin data processing) on 25.0%. These shares add up to more than 100% because a family can carry several IPC classes at once.
Shares measured against the 116-family total
Legal status
62 of 163
published records
62 of 163 records were granted and in force
At the retrieval date, 62 of the 163 published records (38%) were granted and still in force. The remaining records are pending, withdrawn, refused, lapsed or not determinable from this view; that breakdown is not separated out here, and in-force status reflects legal standing only, not technical or commercial significance.
Status measured at 2026-10-15 retrieval
Eureka AI Agent
Looking for what nobody has claimed yet?
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to bayesian network structure learning from data patent landscape, with the prior art for and against each one.
Source: Patsnap Eureka. Derived observations on the same dataset. Derived from one Patsnap search for Bayesian Network Structure Learning from Data Patent Landscape, covering all matching records regardless of filing date (the search string sets no date limit), retrieved 2026-10-15. Counts cover published records only and move as new filings publish.Run this in Eureka MCP
What's Next
Where to take this next
The figures above describe what this search returned. Reading individual claim sets, tracking a specific filer, or rerunning the search with adjusted terms are the natural next steps.
Pull granted claims before drawing scope conclusions
A family count or an abstract cannot establish what any patent's granted claims require. Pulling the full claim set for records of interest is the only way to assess scope.
Open this dataset in Patsnap Eureka →Track filers beyond the top rows
Most of the 90 listed filers in this set hold 2 or fewer families each. Watching how that list changes over time captures shifts a single snapshot cannot.
Explore filer activity in Patsnap Eureka →Re-run the search with adjusted terms
Small IPC branches in this dataset may reflect search-string noise as much as genuine activity. Adjusting the query terms can help separate the two.
Refine this search in Patsnap Eureka →Source: Patsnap Eureka. Suggested next steps. No new data: every figure named here is one of those above. Derived from one Patsnap search for Bayesian Network Structure Learning from Data Patent Landscape, covering all matching records regardless of filing date (the search string sets no date limit), retrieved 2026-10-15. Counts cover published records only and move as new filings publish.Run this in Eureka MCP
FAQ
Common questions about this dataset
This search returns 116 simple patent families, which de-duplicate a single invention across the countries it was filed in, and 163 published records, which count each publication separately. There is no date limit on the underlying search string, so this count spans every matching record the search returns rather than a fixed year range. Because the search string matches on title, abstract and claim text for specific terms, it will miss records that describe the same techniques with different wording, so these counts should be read as one search's result rather than a definitive count of all related patents.
Intel Corp is named on 6 of the 116 simple patent families in scope, a 5.2% share, the largest single-row figure in the filer list returned by this search. That list covers 90 organisations in total, counted in families, with filings by named individuals removed from the list (though still inside the overall totals). Because one family can name more than one filer, these rows cannot be added together to produce a combined concentration figure, and the list itself is not rolled up to corporate parent, so related entities may appear as separate rows.
The data shows a long tail rather than concentration at the top: 84 of the 90 listed filers hold 2 or fewer simple patent families each, while the largest single filer, Intel Corp, holds 6 (5.2% of 116). That pattern suggests many organisations have filed a small number of families rather than a small number of organisations holding most of the field, but filer rows cannot be summed to produce an exact concentration percentage because a family can name multiple filers. The filer list also excludes individual inventors, who are still included in the overall family and record totals.
Measured against the 116 simple patent families in scope, G06N (computing based on AI models) appears on 65.5% of families, G06F (electric digital data processing) on 50.9%, and G06Q (business, commerce & admin data processing) on 25.0%. These three carry the highest shares of the 27 IPC subclasses found across the dataset. Because a single family can carry more than one IPC class, these shares add up to more than 100% and should not be read as a partition of the dataset into exclusive buckets.
The filing trend, counted in published records by earliest filing year, runs from 2 records in 2007 to a peak of 22 in 2021, with 6 so far recorded for 2026. Records filed from 2022 onward are still in the process of publishing, since publication can lag filing by 18 months or more, so the recent years in the chart understate their eventual totals and should not be read as a slowdown. This dataset does not support a stated growth rate or trend direction beyond describing the shape of the chart as given.
This published application, assigned to Huawei Technologies Co., Ltd. and published 2023-02-09, describes a reinforcement learning method and apparatus built around a structure graph: it feeds a current environment state and a learned structure graph into a graph-neural-network policy function, which generates an action, and then observes the resulting next state from the environment. It is the first record returned under this search's default ordering, not a top-cited or most significant filing. What its granted claims actually require, and how broadly they reach, can only be determined by reading the claim set itself rather than the abstract.
Answers are grounded in the same dataset. Derived from one Patsnap search for Bayesian Network Structure Learning from Data Patent Landscape, covering all matching records regardless of filing date (the search string sets no date limit), retrieved 2026-10-15. Counts cover published records only and move as new filings publish.Run this in Eureka MCP
Research Bayesian Network Structure Learning from Data Patent Landscape in depth with Eureka
Go past this page: query the whole bayesian network structure learning from data patent landscape corpus yourself, in your own scope.
Every answer comes back with patent numbers you can open.
How this page was built, and what the numbers do and do not mean
Search. Every figure on this page comes from one Patsnap search, run once and not edited afterwards. This is the exact search string:
TAC:(("bayesian network*") AND ("structure learning" OR "structure discovery") AND (data OR graph*))Reading the field codes: TAC searches the title, abstract and claims of each record, not the full description. The search string sets no date limit, so the scope is every matching record in the database regardless of when it was filed. It is not restricted to recent years. How this string was built matters for how far the counts can be pushed. It was generated in batch from a topic list rather than hand-crafted for this page, and the result set is not manually screened. A keyword search over title, abstract and claims both over-captures (unrelated filings that happen to use the same words) and under-captures (the same invention written in different words). This page cannot measure either effect, so read every figure here as what this particular string returned, not as what exists in the field. One subclass (G06N) carries 65.5% of the patent families, so this search returned a classification-tight set.
What is being counted. Two different units appear on this page and they are not interchangeable. The filer list and the technology split are counted in simple (DOCDB) patent families: one invention counts once however many countries it was filed in. The total, the jurisdiction split and the filing-year chart are counted in published records: the same invention counts once per publication. For this search that is 163 records and 116 families. Record counts reward filing in many jurisdictions; family counts do not. Shares are only ever taken against a denominator in the same unit, and each one names it.
Who the filers are. Names come from Patsnap's standardised assignee field, as recorded on the documents themselves. They are not rolled up to corporate group: a parent and its subsidiaries appear as separate entries, and a portfolio acquired from another company stays under the name on the document unless the assignment is reflected there. This list is meant to carry organisations, not people. Filings held by a named individual are identified by matching the assignee against the inventors on the same search, and, for names written in Chinese, Japanese or Korean characters, by the name carrying no organisational suffix; those are removed from the list (10 of them here). Neither test is exhaustive: the inventor list it matches against is itself capped, and a personal name written in Latin script with no organisational wording in it cannot be told apart from a short company name by rule. So an individually held filing can still appear in the list. Those filings stay inside every total and share on this page either way, so the listed entries can add up to less than the total. A record shown in full further down still names its assignee exactly as the patent office records it, including where that is a person: that name is how the record is identified and checked. What is left out of the ranking is people as ranked entries, not people as cited sources. The table shows the top 30 of the 90 filers returned, ordered by count alone; the data endpoint returns at most 100 entities, so even that 90 is not every filer. One patent family can name several filers and each is counted separately, so these rows are not slices of a divided whole: adding them together over-counts, and this page deliberately publishes no combined "top five" share for that reason. See every filer, individuals included
Dates. The year chart counts by earliest filing year, not publication year. A filing only becomes visible once it publishes: roughly 18 months for most routes, and up to 30 months where a PCT application enters national phase. Measured across these pages, the year four back from today still holds only about 89% of its eventual count. 2022 onwards are therefore shaded on the chart as still filling in. The data behind this page was retrieved on 2026-10-15.
Legal status. "In force" counts documents whose legal status at retrieval was granted and maintained (62 of 163). Everything else is pending, withdrawn, refused, lapsed or not determinable, and those states are not broken out here because the underlying codes do not separate them cleanly. Legal status changes continuously and is reported per document, not per family. Get the status breakdown in Eureka
Classification. The technology split uses IPC subclasses (four characters, for example H01M). A single record usually carries several, so the shares add up to more than 100% and a record can appear in more than one bar. Subclass is a coarse grouping; it describes the field a record was classified into, not what the claims cover. Break the classes down to subgroup level
Who wrote the words. The tables, charts, counts, shares and patent numbers on this page are produced directly from the search result by a program; none of them are written or estimated by a language model. The surrounding prose, headings, FAQ answers and the answers in the question panel are drafted by a language model that is given those same figures and is not permitted to introduce others. Prose can still mis-describe a correct figure, so where the two disagree, the tables and charts are the record.
Disclaimer. This page is generated from Patsnap Eureka data drawn from a limited snapshot of global patent and scientific-literature records, and is provided for general information and reference only.
Patent data carries limitations that apply to every figure above. Recent filings are under-counted, because an application is not public until roughly 18 months after filing and database loading adds further delay. The search string defines and constrains everything shown: records that use different terminology, or that sit outside the fields searched, are absent and cannot be counted. Applicant names, classification codes, legal-status codes and citation records are taken as recorded and may contain errors, duplicates, omissions or stale entries. Ordering by volume is not a measure of quality, scope, enforceability or commercial value, and a larger portfolio is not a stronger one. As a result, the analysis may be incomplete or inaccurate and may not reflect the full technology landscape.
Nothing on this page constitutes an exhaustive prior-art, novelty, freedom-to-operate, validity or infringement search, nor does it constitute legal, financial, investment, or professional advice, and it should not be relied upon as such. No attorney-client relationship is created by it. Any patent, commercial, or strategic decision should be verified independently against the primary records and reviewed with qualified patent, legal, and domain professionals. Patsnap makes no warranties, express or implied, as to the accuracy, completeness, or fitness for any particular purpose of the information presented.
The question panel. The answers behind the suggested questions on this page were written when the page was generated, not in response to your click. The progress steps that play before an answer appears are an animation of how that analysis was done; nothing is being searched or computed at that moment. Typing your own question instead opens Patsnap Eureka, which does run it live.