APOGEE

Cross-Source Intelligence

Capabilities that span multiple data sources - stance tracking, unified entity timelines, policy reversal detection, and issue attention tracking across legislation, hearings, news, lobbying, and press releases.

Cross-Source Intelligence

See the summary view → for a quick overview of all Cross-Source Intelligence capabilities.

4 capabilities that synthesize across multiple data sources. These work at the intersections - combining press releases, hearings, news, lobbying, and legislation to answer questions no single source can.

Position & Stance Tracking

Query stance data extracted from congressional press releases and hearing testimony to understand where members, organizations, and witnesses stand on specific bills and policy issues. Three query directions: by member (what has this member supported or opposed?), by entity (who supports or opposes a specific bill?), and by witness (what positions did witnesses take in hearing testimony?). Filter by stance type (support, oppose, neutral), source type (press release or hearing), and date range.

Stance data is extracted from press releases using AI-powered entity and sentiment extraction, and from hearing transcripts using witness testimony analysis. This makes it possible to search across thousands of press releases and hearing transcripts for concrete position statements - a task that would take weeks of manual reading.

How it works: Press releases are processed through entity extraction that identifies bill mentions and characterizes the member's or organization's stance (support, oppose, neutral). Hearing transcripts are segmented by witness and analyzed for position statements. Both are stored as typed relationships in the knowledge graph and queryable through a unified tool.

Data: 40,547+ stance relationships extracted from congressional press releases, plus witness position data from GPO hearing transcripts. Stances are linked to canonical entities (members, bills, organizations) in the knowledge graph.

What has Senator Warren said about crypto regulation? Try it → Which members have publicly opposed the AI safety bill? Try it → What positions did witnesses take at the hearing on social media and children? Try it → Show me all organizational stances on HR 1 from press releases Try it →

Cross-Source Entity Activity

Unified activity timeline for any member, organization, or bill - pulling from news coverage, press release stances, legislation actions, lobbying filings, and hearing testimony into a single chronological view. Answers "what's been happening with X?" across every data source Apogee tracks, eliminating the need to search six different databases and manually correlate results.

This is particularly valuable for preparing meeting briefings, monitoring client-relevant entities, or catching up on developments you may have missed. Instead of checking news, then lobbying filings, then bill status, then committee activity separately, one query returns a complete activity timeline sorted by date.

How it works: A graph traversal from the target entity (Member, Organization, or Bill) follows all relationship types - news mentions, press release stances, bill actions, lobbying filings, hearing appearances - and returns results as a unified reverse-chronological timeline. Supports entity type filtering and date range limiting.

Data: Knowledge graph combining Congress.gov bill actions, GPO hearing transcripts, SOPR lobbying filings, press release stances, and news coverage from 40+ outlets - all linked through entity resolution to canonical nodes.

What's been happening with the American Medical Association in the past month? Try it → Show me all recent activity for Senator Schumer - bills, news, stances, hearings Try it → Give me a timeline of everything related to the CHIPS Act across all sources Try it →

Policy Reversal Detection

Finds entities that changed their stance on a bill or policy topic - detecting when a member or organization shifted from support to opposition (or vice versa) based on press release stance data. Three analysis modes: by bill (who flipped on this bill?), by member (which bills did this member flip on?), and by topic (search topic-related bills for any reversals).

This capability addresses one of the most important accountability questions in politics: do members follow through on their public positions? Detecting stance reversals early is critical for journalists, advocacy organizations, and government affairs teams tracking legislative commitments.

How it works: Press release stances are analyzed chronologically per entity-bill pair. When an entity's stance on a bill changes over time (e.g., support → oppose), the reversal is flagged with the original and new stance, confidence scores, and timestamps. Topic mode searches bill titles first, then checks all matched bills for reversals.

Data: Press release stance extraction data linked to canonical entities and bills in the knowledge graph. Supports configurable lookback window (default 365 days) and confidence thresholds.

Has any member reversed their position on the AI safety bill? Try it → Which bills did Senator Manchin change his stance on this Congress? Try it → Detect position reversals on climate legislation Try it → Who flipped on the infrastructure bill? Try it →

Issue Attention Tracking

Longitudinal tracking of attention to any entity or policy issue across all data sources simultaneously - news coverage, press release mentions, hearing appearances, and lobbying filings - producing a multi-channel time-series that reveals whether an issue is building sustained momentum or generating only temporary spikes.

While media surge detection identifies short-term spikes in a single source, issue attention tracking shows the longer arc across all sources at once. An issue with rising attention across news, hearings, and lobbying simultaneously is far more likely to drive legislative action than one generating only media coverage.

How it works: Time-series aggregation of entity mentions across news articles, press releases, hearings, and lobbying filings. Returns per-source counts in configurable time buckets (day, week, or month) with summary statistics including total mentions, peak period, trend direction, and percent change. Supports configurable lookback windows up to 365 days.

Data: Entity mention relationships in the knowledge graph spanning news articles (40+ outlets), congressional press releases, GPO hearing transcripts, and SOPR lobbying filings. Updated as new data is ingested.

Is AI regulation getting more or less attention across all sources? Try it → How has attention on Senator Warren trended over the past 6 months? Try it → Which policy issues are building momentum across multiple channels? Try it → Track the Chamber of Commerce's media presence over time Try it →

Ready to try these capabilities? Get started with Apogee or try the chat client.