ScryCX


Data scientists are often faced with selecting the most efficient algorithms to solve a given business problem. This process can be very time-consuming if they do not already know what they are looking for. By combining domain expertise and Big Data Science techniques, the ScryJidoka® platform provides a decision support system (DSS) to help users choose and execute appropriate algorithms in order to analyze their data and solve the problem. To achieve this, ScryJidoka®‘s DSS uses a rich library of statistical and mathematical operators to analyze the settings in which various algorithms were previously successful, thereby providing appropriate recommendations. Some of the highlighted features of ScryJidoka® are given below.

Banking, Financial Services & Insurance

Risk & Compliance Analytics

ScryCredit® analyzes a borrower’s "financial health" to help in making credit decisions and reducing loan defaults. By using internal and external data that is difficult to process manually, it evaluates businesses and consumers with respect to cash flows. ScryCredit® has pre-built machine learning algorithms that have been pre-trained using financial statements and customer sentiment ontology to generate creditworthiness scores, which compliment credit ratings.

Benefits of ScryCredit® Lending and Credit Risk

  • Lower false negatives for potential customers with lower credit rating but strengthening cash flow predictions and end-customer sentiment.
  • Lower false positives for existing borrowers with good credit rating but weak cash flow and end-customer sentiment trends.
  • Cash flow and financial statements ontology to understand transactions and identify potential default situations.
  • Recommendation engine to analyze existing customers’ cash flows to identify and recommend loan prospects based on multiple customizable criteria.
The volume and complexity of payment transactions creates many challenges for banks. Most of them analyze payments manually to look for transactions that could represent any known category of risk. This means they must analyze a very small sample, from a large number of transactions, leaving potential anomalies undetected. Moreover, analysts are constrained to process limited information in detecting anomalies. ScryAnomaly ® deploys an ensemble of algorithms to detect and highlight potential risks in payments and transactions. It processes incremental internal and external datasets that complement transactions data to enhance classification of transactions as anomalies.

Benefits of ScryAnomaly® Payments & Transactions

  • Larger coverage of transactions with automated anomaly detection.
  • Faster detection of anomalies from a large set of transactions.
  • Performs additional checks such as prior convictions, known affiliation to other blacklisted organizations or individuals, and presence in international sanctions lists.
  • Leverages external databases to provide incremental data on transacting entities.
  • Provides visual drill-down reports with workflows to aid analysts in classifying anomalous transactions with higher confidence.
  • Reduces anomaly detection costs while enhancing the productivity of detection processes.
Cash flows carry significant details about a business' financial health. However, analyzing them manually at transaction level to find patterns is not practical. ScryRisk ® Monitoring deploys proprietary machine learning and anomaly detection algorithms to enable lending institutions to evaluate cash flows of prospective SMB customers at the transaction level and discover adverse events. Cash inflow and outflow volume details along with time-stamps assists customers in forecasting SMB cash liquidity and hence creditworthiness. The product has machine learning algorithms pre-trained using financial ledger ontology to identify and classify cash transactions and generate cash ratios that bank relationship managers can use to make credit decisions.

Benefits of ScryRisk® Monitoring

  • Cash flow and financial ledger ontology to understand transactions, learn them with time and identify anomalies and patterns.
  • Recommendation engine to analyze existing customers' cash flows to identify and recommend loan prospects based on multiple customizable criteria.
  • Time-series analytics and Knowledge Graphs to predict SMB customers’ financing needs based on their cash flow patterns, spending behavior and trading relationships.
Complaints resolution and services management are a critical part of customer experience. ScryComply® Customers leverages Scry's ScryCX® customer experience products to provide customer experience solution focused on efficiently categorizing and benchmarking complaints, learning and aiding customer experience representatives in proposing next best action to resolve them. Our solutions use proprietary Text Analytics and Natural Language Processing engines to expedite complaints resolution learning from how similar complaints were handled in the past. The solutions use ScryCX® products deployed with banking, financial services and insurance industry ontologies to help improve compliance in customer servicing.

Benefits of ScryComply® Customers

  • Context based recommendations - Tickets can belong to different categories within the same technology; hence recommendations are domain and topic specific. These recommendations are also ranked on the degree of similarity based on Scry's internal ranking algorithm and achieve 91%+ accuracy.
  • Improved efficiency - Uses advanced ML algorithms to recommend list of solutions based on similar issues/resolutions in the past.
  • Automatic classification and categorization of complaints with our Text Analytics Engine.
  • Historical trends & highlighting complaints that occur frequently; helps reduce number of open items and improves customer experience.
  • Automated Decision Support for "Next Best Action" to resolve new incoming complaints.
  • Identification of trending issues and initiating associated alerts.
  • Comparison & benchmarking of complaint resolution performance across different products.
Customer Prospecting

Less than 25% of SMB and consumer loans are approved at major financial institutions. Furthermore, lengthy credit approval processes and low approval rates increase their loan origination costs. It is expensive and inefficient for bank relationship managers to acquire and manually process information from multiple sources. ScryProspect® uses proprietary AI-based algorithms to accelerate credit approvals by processing structured and unstructured data that is obtained from internal and external data sources, and it provides predictive and prescriptive actionable insights on creditworthiness of an SMB or an end consumer.

Benefits of ScryProspect®

  • Unstructured data processing engine to enrich customer prospecting process by collecting, mining and analyzing consumer data from public data sources such as Google+, BBB, Yelp, news articles and Dun & Bradstreet.
  • Recommendation engine that analyzes existing customers to identify and recommend loan prospects based on multiple customizable criteria.
  • Time-series machine learning algorithms and Knowledge Graphs to predict SMB customers’ financing needs based on their cash flow patterns, spending behavior and trading relationships.
The solution focuses on improving per customer yield and higher sales performance with machine learning and artificial intelligence. It helps form cohorts of your customers based on their behavioral attributes and builds a comprehensive profile for each of your customers to allow targeted marketing strategies or sales tactics. It automatically screens your customers to identify and recommend customers with high upsell and cross-sell potential using proprietary AI-based algorithms to process structured and unstructured data that is obtained from internal and external data sources.

Benefits of ScryProspect®

  • Unstructured data processing engine to enrich customer prospecting process by collecting, mining and analyzing consumer data from public data sources such as Google+, BBB, Yelp, news articles.
  • Recommendation engine that analyzes existing customers to identify and recommend prospects based on multiple customizable criteria.
  • Time-series machine learning algorithms and Knowledge Graphs to predict consumers buying needs from data such as past purchase transactions, product usage, behavioral attributes and social media.
  • Additional cash flows and credit transaction data processed to prospect customers for credit approvals and banking products.

Procurement, Sourcing & Supply Chain

ScryRisk® Vendors helps organizations assess their ecosystem of vendors on an ongoing basis. It analyses vendors' financial and business performance data, structured and unstructured, from third-party data providers, public sources and social media. The vendor business performance score helps organizations proactively mitigate their risks and minimize their exposure to events that may impact vendors.

Benefits of ScryRisk® Vendors

  • Aggregation of data from third-party data providers such as D&B, LexisNexis, Yelp and Facebook.
  • Better understanding of your vendor base through Scry’s vendor analysis models.
  • Valuation based on customer metrics and network effects (who else works with your vendors).
ScryDigitize® Utility Bills is a scalable pre-trained solution to extract data from utility bills that are in pdf or image format and convert it into formats ready for applications.
ScryDigitize® Invoices is a scalable pre-trained solution to extract data from vendor invoices that are in pdf or image format and convert it into formats ready for applications. Combined with ScryAudit®, it can connect with payment systems to validate payments against invoices and highlight audit anomalies.
ScryDigitize® Agreements is a scalable pre-trained solution to extract attributes from rent rolls in pdf or image format and convert it into formats ready for applications.
ScryAudit® offers APIs to connect ScryDigitize® products to internal data sources to validate data across multiple systems. When combined with ScrySSoT® products, ScryAudit® offers intelligent connections between data in internal systems and new data ingested using ScryDigitize® products.

Customer Experience

ScryCX®
Your customers engage with your organization across multiple channels leaving product quality and usage insights
embedded in unstructured-data format in feedback, inquiry emails, communication with contact center, discussion
forums and blogs. ScryCX® products help your organization connect pieces of information in structured and unstructured data across multiple systems to create stronger customer experience.Customer experience products across ScryCX® are:

Services Management teams collect and process large amounts of unstructured data such as emails, log files, and reference material such as product manuals and service tickets history. Processing this unstructured data for service tickets is time and effort intensive. Finding the degree of similarity between a new case and cases from the past is challenging. Scry's Service Management helps expedite ticket resolution by automating the processing of unstructured data and identifying past tickets with a high degree of similarity. The solution uncovers the underlying semantic structure of service request by identifying recurring patterns that can be used to find similarity with tickets solved in the past.

Benefits of ScryCX® Service Management

  • Context based recommendations - Tickets can belong to different categories within the same technology; hence recommendations are domain and topic specific. These recommendations are also ranked on the degree of similarity based on Scry's internal ranking algorithm and achieve 91%+ accuracy.
  • Improved efficiency - Uses advanced ML algorithms to recommend list of solutions based on similar issues/resolutions in the past.
  • Lower Cost – Reduction in time and cost involved in solving each service ticket; reduces number of open issues leading to higher customer satisfaction.
Complaints resolution is a critical part of customer experience. Scry's Complaints Management solution focuses on efficiently categorizing and benchmarking complaints, learning and aiding customer experience representatives in proposing next best action to resolve them. Our solution uses proprietary Text Analytics and Natural Language Processing engines to expedite complaints resolution learning from how similar complaints were handled in the past.

Benefits of ScryCX® Complaints Management

  • Automatic classification and categorization of complaints with our Text Analytics Engine.
  • Identification of trending issues and initiating associated alerts.
  • Historical trends & highlighting complaints that occur frequently; helps reduce number of open items and improves customer experience.
  • Comparison & benchmarking of complaint resolution performance across different products.
  • Automated Decision Support for “Next Best Action” to resolve new incoming complaints.
ScryCX® provides contact center operational metrics, adding intelligence to process this data for improved business outcomes. Calls are automatically classified into categories such as product type, customer segment or machine-learned clusters. The solution processes structured business data, unstructured customer conversation and reference data, and agent notes to make call rerouting and call resolution decisions ensuring that the agent has the next best action for the customer interaction.

Benefits of ScryCX® Contact Center Analytics

  • Leverages Data Quality (DQ) and Business Rules (BR) Engines of ScryCollatio® in addition to connectors and scrapers libraries to ingest unstructured data.
  • Provides descriptive, predictive and prescriptive analytics on contact center operational data.
  • Offers capability to consume unstructured data from multiple third-party databases and public sources.
Compliance with processes is significant for Contact Centers. However, how do they ensure compliance without introducing burden of additional processes and manual monitoring? For example, agents may not write the correct reasons for delinquency of a customer choosing between inability to pay and unwillingness to pay. Validating compliance of agents' action is time consuming and laborious, and only 2% of calls are reviewed. Scry’s solution deploys speech-to-text analytics for compliance checks.

Benefits of ScryCX® Contact Center Compliance

  • Automated categorization of reasons for delinquency.
  • Recommendations based on the degree of similarity by Scry's cognitive computing solution with 90%+ accuracy.
  • Solution checks all contact center conversation and provides "percentage of compliance" for each conversation.
  • Capable of running real-time by analyzing limited conversations with speech-to-text, NLP and ML.

ScryInnovate® Products

Scry Analytics builds intellectual property jointly with clients, productizing new ideas – ours or yours. We do it faster and better by using Scry’s ML/AI building blocks:

  • ScryJidoka® – proprietary and open source ML/NLP algorithms
  • ScryCollatio® – Proprietary data-engineering platform

We bring in efficiency through reusable connectors, scrapers and APIs to bring in external data – structured and unstructured. We understand industries and therefore build domain specific ontologies & definition directories using our business domain and data-science expertise. The outcome are products with strong algorithms and rich user interface with knowledge graphs and hypergraphs.

Oncology & Hematology

Social and online health forums allow people with similar health conditions to interact with each other and share experiences. This is real patient data and it holds vast amount of untapped information. However this data is unstructured and spread across hundreds of sites and millions of messages, which makes it incomprehensible to derive any reliable and intelligent insight. VoCP brings all these conversation from different sites and forums onto a single platform and helps patients, healthcare professionals and pharma companies to uncover health insights from these real life experiences. Using Natural Language Processing and advanced Machine Learning, VoCP processes millions of conversations and identifies various concepts, entities and relationships of importance.

VoCP helps provide information about:

  • Drugs and medications that cancer patients are taking.
  • Various side-effects that patients are experiencing.
  • Supportive treatments that patients are discussing for alleviating side effects.
  • Perceived effectiveness and sentiment towards particular drug or treatment.
  • Factors discussed by patients affecting their quality of life and ways to overcome them.
  • Data from over 11 million messages from 100s of health forums and social sites.
  • An intuitive platform to connect effectively with other patients to learn and discover.
  • Articles with research and deep insights on various topics from real patient conversations.

 

VISUALIZATION AND DESCRIPTIVE ANALYTICS

To use ScryJidoka®, users input the processed (cleansed and harmonized) data, via Scry-Collatio or otherwise, and use its Graphical User Interface (GUI) to visually understand it. Given this data and the specific business problem, this platform filters for useful algorithms that may be proprietary or Open Source and stored in its library. Through the GUI, the users can see the utility of these algorithms for descriptive and statistical analysis, and then test their hypothesis using any of these algorithms. Finally, based on the outcome, users can either select algorithms individually or use ScryJidoka®‘s DSS to combine them and achieve better results.

 

PREDICTIVE AND PRESCRIPTIVE ANALYTICS

Using the given data, ScryJidoka® helps users determine the parameters that are most relevant to the business problem. Based on these results, this platform predicts the dynamics of the key performance indicators (KPIs) for this problem. Finally, this platform provides prescriptive analysis and actionable insights as to how to improve these KPIs, thereby leading to enhanced performance.

 

OPEN SOURCE AND PROPRIETARY ALGORITHMS

The algorithms – both Open Source and proprietary – stored in Scry-Jidoka’s library are state-of-the-art. In case Open Source algorithms do not meet our needs, we create our own proprietary algorithms to solve specific problems; our algorithms incorporate concepts that are still actively under research by theoretical computer scientists including deep learning and hypergraph modeling. Furthermore, this platform contains many modified algorithms for analyzing unstructured data, audio and video data, and wave-forms data (e.g., ECG data).

 

IMPORTING DATA

ScryJidoka® starts by ingesting different data-sets that may not reside in a single comprehensive database. This data is input either from the Scry-Collatio platform or from clients’ databases (that contain analysis-ready data) through Scry-Collatio’s data-connector tools and application program interfaces. In both situations, cleansing, munging and harmonizing of data has occurred before it is provided to the ScryJidoka® platform.

 

MODELING USING STATISTICS ALGORITHMS

ScryJidoka® platform has pre-built libraries that are used for “exploring” data with respect to potential problems and noise; these software libraries help a user in imputing missing values, computing caps and floors, transforming data (as required for the business problem), and noise reduction. In addition, ScryJidoka® contains several libraries containing statistical algorithms (e.g., generalized linear model, logistic regression, random forests, ARIMA, SARIMA, ETS, Kalman Filters, and related techniques) that help domain experts and data scientists in a deeper understanding of the harmonized data and in deriving actionable insights. For a given business problem, Scry-Jidoka’s decision support system (DSS) helps users choose and execute appropriate algorithms – as well as combine these algorithms if needed – so as to produce better insights. Intuition from data scientists and subject matter experts complements this entire process, thereby, giving this platform its name:
automated computing with a human touch.

 

MACHINE LEARNING & INFORMATION RETRIEVAL ALGORITHMS

The machine learning, natural language processing and information retrieval libraries in ScryJidoka® are designed to handle various steps, which include tokenizing, POS tagging, sentence splitting, entity recognition, relation extraction, feature extraction, topic modeling, similarity algorithms, k-nearest neighbor, state space modeling, and classification algorithms.  In addition, this platform contains our proprietary algorithms that are related to deep learning algorithms and recurrent neural networks, which have also been modified to work extremely well in specific domains (e.g., Oncology) and within specific industries.

 

PRESENTATION AND OUTPUT

For solving a specific problem, data scientists and subject matter experts work together to pick the appropriate algorithms (as suggested by ScryJidoka®‘s DSS or on their own), execute these algorithms on the given data, and obtain the final results. In fact, using ScryJidoka®, these professionals can compare two or more solutions by executing two or more sets of algorithms on the same data set and decide as to which one works better for their use-case. ScryJidoka®‘s visualization provides Graphics User Interface (UI) functionalities so that outputs can be visualized by concurrent users. Indeed, these outputs can be depicted in different kinds of pie-charts, bar-charts, two dimensional graphs, graphs with nodes and edges that depict dataset mapping, data quality distribution, and various object relationships.