cs.IRMar 12, 2026

Behavioral Intelligence Platforms: From Event Streams to Autonomous Insight via Probabilistic Journey Graphs, Behavioral Knowledge Extraction, and Grounded Language Generation

Authors: Arun Patra, Bhushan Vadgave

Organizations: Journium, Inc.

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Contemporary product analytics systems require users to pose explicit queries, such as writing SQL, configuring dashboards, or constructing funnels, before insights can surface. This pull-based paradigm creates a bottleneck: it requires both domain knowledge and technical fluency, and assumes practitioners know in advance which questions to ask. We argue that behavioral analytics should move from passive systems that answer queries to active systems that continuously detect and explain behavioral phenomena. We present the Behavioral Intelligence Platform (BIP), a system architecture that transforms raw event streams into automatically generated insights. BIP consists of four layers. First, Normalization and State Derivation (NSD) standardizes events and maps them to a semantic state hierarchy. Second, a Behavioral Graph Engine (BGE) models user journeys as absorbing Markov chains and computes transition probabilities, removal effects, and path quality metrics. Third, a Behavioral Knowledge Graph (BKG) and Detector System convert graph outputs into grounded behavioral facts and identify behavioral phenomena. Finally, a Grounded Language Layer constrains large language model outputs to verified facts, producing reliable narrative insights. We formalize the Behavioral Intelligence Problem, introduce a taxonomy of detectors for autonomous insight generation, and propose an interestingness score to prioritize insights under limited attention.

Figures & tables

Figure 1: BIP four-layer data-flow architecture. Each layer produces a structured artifact consumed by the next: NSD derives a semantic state hierarchy from raw events; BGE builds absorbing Markov chain snapshots; BKG+DS reify graph metrics into grounded facts and emit typed findings; GLL generates faithful narrative insights constrained to verified facts.
Figure 2: Example six-state funnel as an absorbing Markov chain. Single-bordered circles are transient states ( S\mathcal{S} ); double-bordered circles are absorbing states ( A\mathcal{A} ). Edge labels are empirical transition probabilities; each row sums to 1 across all outgoing edges (thin arrows indicate low-probability direct conversions). The dashed back-edge represents re-visitation from feature_used to sign_up .
Figure 3: Absorption probabilities B[s,converted]B[s,\text{converted}] (Eq. ( 4 )) computed by closed form from the transition matrix in Figure 2 via B=N⋅RB=N\cdot R . Each bar is the probability that a journey starting from state ss ultimately reaches converted . Reproduced exactly by markov_journey_model.py ; the empirical conditional P(converted∣reached(s))P(\text{converted}\mid\text{reached}(s)) from Monte Carlo trajectories agrees with B[s,converted]B[s,\text{converted}] to within sampling noise (see simulate_trajectories.py ), as required by the strong Markov property when all journeys share a common starting state.
Figure 4: Removal effect ranking (Eq. ( 8 )) for the example funnel. Each bar shows the decrease in overall P(converted)P(\text{converted}) when the corresponding state is removed from the journey graph and transition probabilities are re-normalised. import_data is the structurally most critical state, as it is the sole gateway to invite_teammate — the highest-converting transient state.
RankIDDetectorFinding (summary)Score
1F002ActivationDriverfeature_used is activation driver for converted (lift 4.2 ×\times )0.845
2F001TemporalRegressionemail_verified →\to profile_complete drop-off ++ 18 pp post-v2.30.736
3F004SegmentDivergenceMobile conversion 22 pp below desktop0.602
4F003UnexpectedLoopLoop: profile_complete →\to sign_up (8% of users)0.252
5F005DropOffClusterDrop-off at sign_up →\to abandoned (low significance, N=85N{=}85 )0.054
Table 1: Composite interestingness scores for five synthetic detector findings (weights: α=0.30\alpha{=}0.30 , β=0.25\beta{=}0.25 , γ=0.20\gamma{=}0.20 , ω=0.15\omega{=}0.15 , ε=0.10\varepsilon{=}0.10 ). Scores computed from synthetic simulation data by interestingness_scoring.py .
More in this topic

User Modeling

User modeling represents people's interests, knowledge, goals, and changing behavior to personalize predictions and AI assistance.

See all 151 papers

Explore similar work

CardsList
May 26, 2026

Discovery Agents for Real-Time Analytics: Toward Proactive Insight Systems

Modern analytics systems are fundamentally reactive, requiring users to define queries over increasingly complex and continuously evolving data. In real-time streaming environments, this paradigm breaks down, as the space of potential insights becomes too large to enumerate manually. We present a multi-agent architecture for autonomous insight discovery over real-time data streams. The system implements a continuous discovery loop in which agents generate hypotheses, compile them into executable analytics, validate generated artifacts, and produce visualizations and deployable applications. The architecture leverages Apache Kafka for event-driven coordination, Apache Flink for stream processing, and large language models to implement specialized agents. A key contribution is a contract-driven design based on typed intermediate artifacts, enabling modularity, observability, lineage, and safer execution of dynamically generated analytics. Through use cases in retail, finance, and public data, we show how this architecture supports a shift from query-driven analytics to proactive, discovery-driven systems.

Gaetano Rossiello, Dharmashankar Subramanian

IBM · New York City, New York, USA · Yorktown Heights, New York, USA

Multi-Agent LLM SystemsLLM Agent Orchestration
AskPDF
May 8, 2026

Towards Autonomous Business Intelligence via Data-to-Insight Discovery Agent

Transforming fragmented enterprise data into actionable insights remains a significant challenge for LLMs, constrained by complex database schemas, limitations in dynamic SQL generation, and the need for deep multi-dimensional analysis.In this paper, we propose AIDA(Autonomous Insight Discovery Agent), the first end-to-end framework designed for autonomous exploration in complex business environments. We establish a highly flexible instant retail environment encompassing 200+ metrics and 100+ dimensions, and integrates a proprietary Domain-Specific Language (DSL) that bridges semantic reasoning with precise SQL execution. Our reinforcement learning system subsequently formulates business analysis as a Pareto Principle-guided cumulative reasoning process. Experimental results demonstrate that AIDA significantly outperforms workflow-based agents, and extensive evaluations further reveal that AIDA achieves superior environmental perception and more in-depth analysis from diverse perspectives. Our work ultimately establishes the transformative potential of autonomous intelligence for industrial-scale business intelligence systems.

Dongming Wu, Junwen Li, Ming Lu +2

1Rajax Network Technology(Taobao Shangou of Alibaba)

AI-Assisted Decision MakingRL for LLM Agents
AskPDF
Jun 14, 2026

IBAD: Interpretable Behavioral Anomaly Detection on Human Mobility Data

Human mobility appears highly diverse, yet much of a person's daily mobility can be explained by a small set of recurring behavioral templates, such as commuting, school-centered activities, caregiving, nightlife, or errand patterns. We present IBAD (Interpretable Behavioral Anomaly Detection), a framework that learns interpretable daily mobility templates and represents each individual as a distribution over mixtures of these templates. Rather than focusing on specific locations, IBAD characterizes activities that individuals perform across locations. This approach first discovers global behavioral templates using Latent Dirichlet Allocation (LDA), then employs a hierarchical self-supervised model to learn normal behavior of individuals from their soft behavioral templates. We also introduce a splicing benchmark that creates controlled behavioral mismatches between an individual's historical profile and injected mobility patterns. Experiments on real-world and synthetic datasets show that daily behavior can be effectively decomposed into a small number of interpretable templates. Crucially, we show that the learned behavioral archetypes transfer across distinct geographic and demographic contexts. Furthermore, IBAD maintains a robust competitive performance across all settings. For reproducibility purposes, the code is accessible at ~https://github.com/USC-InfoLab/IBAD.

Bita Azarijoo, John Krumm, Cyrus Shahabi

University of Southern California Los Angeles, California, USA

Interpretable Anomaly DetectionAnomaly Detection
AskPDF