Prediction of crime rate

        A new computer model uses publicly available data to predict crime accurately in eight cities in the U.S., while revealing increased police response in wealthy neighborhoods at the expense of less advantaged areas.

        University of Chicago data and social scientists have developed a new algorithm that forecasts crime by learning patterns in time and geographic locations from public data on violent and property crimes. It has demonstrated success at predicting future crimes one week in advance with approximately 90% accuracy.

        The new tool was tested and validated using historical data from the City of Chicago around two broad categories of reported events: violent crimes (homicides, assaults, and batteries) and property crimes (burglaries, thefts, and motor vehicle thefts). 

* OLD APPROACH

        Previous efforts at crime prediction often use an epidemic or seismic approach, where crime is depicted as emerging in “hotspots” that spread to surrounding areas. These tools miss out on the complex social environment of cities, however, and don’t consider the relationship between crime and the effects of police enforcement.

* NEW APPROACH

        The new model isolates crime by looking at the time and spatial coordinates of discrete events and detecting patterns to predict future events. It divides the city into spatial tiles roughly 1,000 feet across and predicts crime within these areas instead of relying on traditional neighborhood or political boundaries, which are also subject to bias. The model performed just as well with data from seven other U.S. cities: Atlanta, Austin, Detroit, Los Angeles, Philadelphia, Portland, and San Francisco.

        The study was supported by the Defense Advanced Research Projects Agency and the Neubauer Collegium for Culture and Society. Additional authors include Victor Rotaru, Yi Huang, and Timmy Li from the University of Chicago. 

#crimes #crimeandpunishment #ML #safeplace

References:

• https://vciba.springeropen.com/articles/10.1186/s42492-021-00075-z
• https://www.sciencedaily.com/releases/2022/06/220630114501.htm
• https://towardsdatascience.com/crime-forecasting-8f71364f2fee

The most profitable area to implement machine learning systems

     Using machine learning for marketing is specifically useful when creating a personalization strategy, with Netflix and Amazon being perfect examples of such practice. Allowing for hyper-personalization algorithms can generate tailored product recommendations based on vast sets of data on consumer behavior, resulting in a substantial increase in sales conversion and customer retention rates so there is no surprise that sales and marketing are the best areas to implement such systems.

    According to Statista, a third of IT decision-makers are interested in rolling out machine learning for business analytics. While a quarter of them look to enhance their security processes, 16% of IT directors use machine learning for marketing. Only 10% of leaders looking to adopt ML do so for customer service purposes.

    

    This is also encouraged by the fact that 80% of machine learning companies target ecommerce and retail industries according to Emerj. Machine learning companies favor ecommerce and retail industries the most, followed by online and social media. The commercial use of machine learning is booming in these industries due to the constant creation of quantifiable data and its relatively easy streaming and storage.

References:

• https://www.statista.com/chart/16951/machine-learning-adoption/
• https://emerj.com/ai-market-research/machine-learning-marketing/
• https://www.mckinsey.com/~/media/mckinsey/featured%20insights/artificial%20intelligence/notes%20from%20the%20ai%20frontier%20applications%20and%20value%20of%20deep%20learning/notes-from-the-ai-frontier-insights-from-hundreds-of-use-cases-discussion-paper.ashx

Hashimoto Thyroiditis with Hypothyroidism Based on Clinical and Paraclinical Data in Female Patients

 

Hashimoto’s thyroiditis is the most common autoimmune disorder and the leading cause of hypothyroidism in iodine-sufficient areas. [1] Untreated hypothyroidism is associated with an increased risk of developing cardiovascular disease, distal polyneuropathy, depression, and myxedema. In recent years, a concept emerged, that thyroid autoimmunity could be associated with low-grade chronic inflammation, independent of thyroid function. It is, therefore, essential to diagnose Hashimoto thyroiditis as early as possible and to test for thyroid function.

Based on these observations, we will try to integrate multiple clinical and paraclinical data, in order to establish an early diagnosis and to elaborate risk prediction of Hashimoto thyroiditis and diverse classification processes, using various machine learning algorithms.

In the last years, multiple studies have tried to identify thyroid disorders early in the course of the disease, based upon the use of hormonal parameters and personal data of the patients, such as gender and age. Some of these published studies use machine learning classification and prediction models, while other studies deal with deep neural network models. The authors, Izdihar and Bozkus, [2] have developed a machine-learning tool for the diagnosis of thyroid diseases, that has the ability to make an intelligent forecast of thyroid gland diseases. Their work showed an overall accuracy of 100% for training and 98.7% for testing. They used decision tree model and support vector machines to help the doctor to label a new case using different parameters.

In our future study, about machine learning model comparison using data about Hashimoto thyroiditis, we will try to find clinical parameters, that demonstrate a significant positive correlation with Hashimoto thyroiditis. For example, age, body mass index, familial history of autoimmune thyroid disease, history of breast cancer, surgically induced menopause, diabetes mellitus type 2, and polycystic ovary syndrome.

Concerning the paraclinical parameters, we will try to see if thyroid autoantibodies, dyslipidemia, uric acid levels, fasting hyperglycemia, and abnormal liver function tests, in non-alcoholic fatty liver disease, could be associated with Hashimoto thyroiditis with hypothyroidism.

 

References

[1] Caturegli, P., De Remigis, A., & Rose, N. R. (2014). Hashimoto thyroiditis: clinical and diagnostic criteria. Autoimmunity reviews, 13(4-5), 391-397.

[2] Al-muwaffaq, I., & Bozkus, Z. (2016). MLTDD: use of machine learning techniques for diagnosis of thyroid gland disorder. Comput Sci Inf Technol, 67-3.

 

Bine ati venit pe blogul cursului de Invatare Automata !